fix conflicts

.gitignore

@@ -24,4 +24,8 @@ output/
 
 build/
 dist/
-paddleocr.egg-info/
\ No newline at end of file
+paddleocr.egg-info/
+/deploy/android_demo/app/OpenCV/
+/deploy/android_demo/app/PaddleLite/
+/deploy/android_demo/app/.cxx/
+/deploy/android_demo/app/cache/

MANIFEST.in

-include LICENSE.txt
+include LICENSE
 include README.md
 
-recursive-include ppocr/utils *.txt utility.py logging.py
-recursive-include ppocr/data/ *.py
+recursive-include ppocr/utils *.txt utility.py logging.py network.py
+recursive-include ppocr/data *.py
 recursive-include ppocr/postprocess *.py
 recursive-include tools/infer *.py
-recursive-include ppocr/utils/e2e_utils/ *.py
\ No newline at end of file
+recursive-include ppocr/utils/e2e_utils *.py
+recursive-include ppstructure *.py
\ No newline at end of file

PPOCRLabel/PPOCRLabel.py

@@ -27,7 +27,12 @@ import json
 import cv2
 
 
+
 __dir__ = os.path.dirname(os.path.abspath(__file__))
+
+import numpy as np
+
+
 sys.path.append(__dir__)
 sys.path.append(os.path.abspath(os.path.join(__dir__, '../..')))
 sys.path.append("..")
@@ -78,7 +83,7 @@ class WindowMixin(object):
             addActions(menu, actions)
         return menu
 
-    def toolbar(self, title, actions=None):  
+    def toolbar(self, title, actions=None):
         toolbar = ToolBar(title)
         toolbar.setObjectName(u'%sToolBar' % title)
         # toolbar.setOrientation(Qt.Vertical)
@@ -92,13 +97,13 @@ class WindowMixin(object):
 class MainWindow(QMainWindow, WindowMixin):
     FIT_WINDOW, FIT_WIDTH, MANUAL_ZOOM = list(range(3))
 
-    def __init__(self, lang="ch", defaultFilename=None, defaultPrefdefClassFile=None, defaultSaveDir=None):
+    def __init__(self, lang="ch", gpu=False, defaultFilename=None, defaultPrefdefClassFile=None, defaultSaveDir=None):
         super(MainWindow, self).__init__()
         self.setWindowTitle(__appname__)
 
         # Load setting in the main thread
         self.settings = Settings()
-        self.settings.load()  
+        self.settings.load()
         settings = self.settings
         self.lang = lang
         # Load string bundle for i18n
@@ -108,7 +113,7 @@ class MainWindow(QMainWindow, WindowMixin):
         getStr = lambda strId: self.stringBundle.getString(strId)
 
         self.defaultSaveDir = defaultSaveDir
-        self.ocr = PaddleOCR(use_pdserving=False, use_angle_cls=True, det=True, cls=True, use_gpu=False, lang=lang)
+        self.ocr = PaddleOCR(use_pdserving=False, use_angle_cls=True, det=True, cls=True, use_gpu=gpu, lang=lang)
 
         if os.path.exists('./data/paddle.png'):
             result = self.ocr.ocr('./data/paddle.png', cls=True, det=True)
@@ -159,7 +164,7 @@ class MainWindow(QMainWindow, WindowMixin):
         filelistLayout = QVBoxLayout()
         filelistLayout.setContentsMargins(0, 0, 0, 0)
         filelistLayout.addWidget(self.fileListWidget)
-        
+
         self.AutoRecognition = QToolButton()
         self.AutoRecognition.setToolButtonStyle(Qt.ToolButtonTextBesideIcon)
         self.AutoRecognition.setIcon(newIcon('Auto'))
@@ -176,7 +181,7 @@ class MainWindow(QMainWindow, WindowMixin):
         self.filedock.setObjectName(getStr('files'))
         self.filedock.setWidget(fileListContainer)
         self.addDockWidget(Qt.LeftDockWidgetArea, self.filedock)
-        
+
         ######## Right area ##########
         listLayout = QVBoxLayout()
         listLayout.setContentsMargins(0, 0, 0, 0)
@@ -250,7 +255,7 @@ class MainWindow(QMainWindow, WindowMixin):
         self.imgsplider.setMaximum(150)
         self.imgsplider.setSingleStep(1)
         self.imgsplider.setTickPosition(QSlider.TicksBelow)
-        self.imgsplider.setTickInterval(1) 
+        self.imgsplider.setTickInterval(1)
         op = QGraphicsOpacityEffect()
         op.setOpacity(0.2)
         self.imgsplider.setGraphicsEffect(op)
@@ -266,7 +271,9 @@ class MainWindow(QMainWindow, WindowMixin):
         self.zoomWidget = ZoomWidget()
         self.colorDialog = ColorDialog(parent=self)
         self.zoomWidgetValue = self.zoomWidget.value()
-        
+
+        self.msgBox = QMessageBox()
+
         ########## thumbnail #########
         hlayout = QHBoxLayout()
         m = (0, 0, 0, 0)
@@ -294,7 +301,7 @@ class MainWindow(QMainWindow, WindowMixin):
         self.nextButton.setStyleSheet('border: none;')
         self.nextButton.clicked.connect(self.openNextImg)
         self.nextButton.setShortcut('d')
-        
+
         hlayout.addWidget(self.preButton)
         hlayout.addWidget(self.iconlist)
         hlayout.addWidget(self.nextButton)
@@ -303,7 +310,7 @@ class MainWindow(QMainWindow, WindowMixin):
         iconListContainer = QWidget()
         iconListContainer.setLayout(hlayout)
         iconListContainer.setFixedHeight(100)
-        
+
         ########### Canvas ###########
         self.canvas = Canvas(parent=self)
         self.canvas.zoomRequest.connect(self.zoomRequest)
@@ -360,6 +367,9 @@ class MainWindow(QMainWindow, WindowMixin):
         opendir = action(getStr('openDir'), self.openDirDialog,
                          'Ctrl+u', 'open', getStr('openDir'))
 
+        open_dataset_dir = action(getStr('openDatasetDir'), self.openDatasetDirDialog,
+                         'Ctrl+p', 'open', getStr('openDatasetDir'), enabled=False)
+
         save = action(getStr('save'), self.saveFile,
                       'Ctrl+V', 'verify', getStr('saveDetail'), enabled=False)
 
@@ -398,6 +408,7 @@ class MainWindow(QMainWindow, WindowMixin):
         help = action(getStr('tutorial'), self.showTutorialDialog, None, 'help', getStr('tutorialDetail'))
         showInfo = action(getStr('info'), self.showInfoDialog, None, 'help', getStr('info'))
         showSteps = action(getStr('steps'), self.showStepsDialog, None, 'help', getStr('steps'))
+        showKeys = action(getStr('keys'), self.showKeysDialog, None, 'help', getStr('keys'))
 
         zoom = QWidgetAction(self)
         zoom.setDefaultWidget(self.zoomWidget)
@@ -438,7 +449,7 @@ class MainWindow(QMainWindow, WindowMixin):
         AutoRec = action(getStr('autoRecognition'), self.autoRecognition,
                       '', 'Auto', getStr('autoRecognition'), enabled=False)
 
-        reRec = action(getStr('reRecognition'), self.reRecognition, 
+        reRec = action(getStr('reRecognition'), self.reRecognition,
                       'Ctrl+Shift+R', 'reRec', getStr('reRecognition'), enabled=False)
 
         singleRere = action(getStr('singleRe'), self.singleRerecognition,
@@ -456,6 +467,12 @@ class MainWindow(QMainWindow, WindowMixin):
         undoLastPoint = action(getStr("undoLastPoint"), self.canvas.undoLastPoint,
                                'Ctrl+Z', "undo", getStr("undoLastPoint"), enabled=False)
 
+        rotateLeft = action(getStr("rotateLeft"), partial(self.rotateImgAction,1),
+                               'Ctrl+Alt+L', "rotateLeft", getStr("rotateLeft"), enabled=False)
+
+        rotateRight = action(getStr("rotateRight"), partial(self.rotateImgAction,-1),
+                               'Ctrl+Alt+R', "rotateRight", getStr("rotateRight"), enabled=False)
+
         undo = action(getStr("undo"), self.undoShapeEdit,
                       'Ctrl+Z', "undo", getStr("undo"), enabled=False)
 
@@ -519,13 +536,14 @@ class MainWindow(QMainWindow, WindowMixin):
                               zoom=zoom, zoomIn=zoomIn, zoomOut=zoomOut, zoomOrg=zoomOrg,
                               fitWindow=fitWindow, fitWidth=fitWidth,
                               zoomActions=zoomActions, saveLabel=saveLabel,
-                              undo=undo, undoLastPoint=undoLastPoint,
+                              undo=undo, undoLastPoint=undoLastPoint,open_dataset_dir=open_dataset_dir,
+                              rotateLeft=rotateLeft,rotateRight=rotateRight,
                               fileMenuActions=(
-                                  opendir, saveLabel,  resetAll, quit),
+                                  opendir,  open_dataset_dir, saveLabel,  resetAll, quit),
                               beginner=(), advanced=(),
                               editMenu=(createpoly, edit, copy, delete,singleRere,None, undo, undoLastPoint,
-                                        None, color1, self.drawSquaresOption),
-                              beginnerContext=(create, edit, copy, delete, singleRere),
+                                        None, rotateLeft, rotateRight, None, color1, self.drawSquaresOption),
+                              beginnerContext=(create, edit, copy, delete, singleRere, rotateLeft, rotateRight,),
                               advancedContext=(createMode, editMode, edit, copy,
                                                delete, shapeLineColor, shapeFillColor),
                               onLoadActive=(
@@ -563,9 +581,9 @@ class MainWindow(QMainWindow, WindowMixin):
         self.autoSaveOption.triggered.connect(self.autoSaveFunc)
 
         addActions(self.menus.file,
-                   (opendir, None, saveLabel, saveRec, self.autoSaveOption, None, resetAll, deleteImg, quit))
+                   (opendir, open_dataset_dir, None, saveLabel, saveRec, self.autoSaveOption, None, resetAll, deleteImg, quit))
 
-        addActions(self.menus.help, (showSteps, showInfo))
+        addActions(self.menus.help, (showKeys,showSteps, showInfo))
         addActions(self.menus.view, (
             self.displayLabelOption, self.labelDialogOption,
              None,
@@ -760,6 +778,10 @@ class MainWindow(QMainWindow, WindowMixin):
         msg = stepsInfo(self.lang)
         QMessageBox.information(self, u'Information', msg)
 
+    def showKeysDialog(self):
+        msg = keysInfo(self.lang)
+        QMessageBox.information(self, u'Information', msg)
+
     def createShape(self):
         assert self.beginner()
         self.canvas.setEditing(False)
@@ -773,6 +795,38 @@ class MainWindow(QMainWindow, WindowMixin):
         self.actions.create.setEnabled(False)
         self.actions.undoLastPoint.setEnabled(True)
 
+    def rotateImg(self, filename, k, _value):
+
+        self.actions.rotateRight.setEnabled(_value)
+        pix = cv2.imread(filename)
+        pix = np.rot90(pix, k)
+        cv2.imwrite(filename, pix)
+        self.canvas.update()
+        self.loadFile(filename)
+
+    def rotateImgWarn(self):
+        if self.lang == 'ch':
+            self.msgBox.warning (self, "提示", "\n 该图片已经有标注框,旋转操作会打乱标注,建议清除标注框后旋转。")
+        else:
+            self.msgBox.warning (self, "Warn", "\n The picture already has a label box, and rotation will disrupt the label.\
+             It is recommended to clear the label box and rotate it.")
+
+    def rotateImgAction(self, k=1, _value=False):
+
+        filename = self.mImgList[self.currIndex]
+
+        if os.path.exists(filename):
+            if self.itemsToShapesbox:
+                self.rotateImgWarn()
+            else:
+                self.saveFile()
+                self.dirty = False
+                self.rotateImg(filename=filename, k=k, _value=True)
+        else:
+            self.rotateImgWarn()
+            self.actions.rotateRight.setEnabled(False)
+            self.actions.rotateLeft.setEnabled(False)
+
     def toggleDrawingSensitive(self, drawing=True):
         """In the middle of drawing, toggling between modes should be disabled."""
         self.actions.editMode.setEnabled(not drawing)
@@ -880,7 +934,12 @@ class MainWindow(QMainWindow, WindowMixin):
             self.updateComboBox()
 
     def updateBoxlist(self):
-        for shape in self.canvas.selectedShapes+[self.canvas.hShape]:
+        self.canvas.selectedShapes_hShape = []
+        if self.canvas.hShape != None:
+            self.canvas.selectedShapes_hShape = self.canvas.selectedShapes + [self.canvas.hShape]
+        else:
+            self.canvas.selectedShapes_hShape = self.canvas.selectedShapes
+        for shape in self.canvas.selectedShapes_hShape:
             item = self.shapesToItemsbox[shape]  # listitem
             text = [(int(p.x()), int(p.y())) for p in shape.points]
             item.setText(str(text))
@@ -1239,6 +1298,8 @@ class MainWindow(QMainWindow, WindowMixin):
 
     def loadFile(self, filePath=None):
         """Load the specified file, or the last opened file if None."""
+        if self.dirty:
+            self.mayContinue()
         self.resetState()
         self.canvas.setEnabled(False)
         if filePath is None:
@@ -1267,7 +1328,7 @@ class MainWindow(QMainWindow, WindowMixin):
                         titem = self.iconlist.item(i)
                         titem.setSelected(True)
                         self.iconlist.scrollToItem(titem)
-                        break 
+                        break
             else:
                 self.fileListWidget.clear()
                 self.mImgList.clear()
@@ -1275,7 +1336,7 @@ class MainWindow(QMainWindow, WindowMixin):
 
         # if unicodeFilePath and self.iconList.count() > 0:
         #     if unicodeFilePath in self.mImgList:
-                
+
         if unicodeFilePath and os.path.exists(unicodeFilePath):
             self.canvas.verified = False
 
@@ -1306,7 +1367,7 @@ class MainWindow(QMainWindow, WindowMixin):
             self.addRecentFile(self.filePath)
             self.toggleActions(True)
             self.showBoundingBoxFromPPlabel(filePath)
-            
+
             self.setWindowTitle(__appname__ + ' ' + filePath)
 
             # Default : select last item if there is at least one item
@@ -1318,7 +1379,7 @@ class MainWindow(QMainWindow, WindowMixin):
             return True
         return False
 
-    
+
     def showBoundingBoxFromPPlabel(self, filePath):
         imgidx = self.getImglabelidx(filePath)
         if imgidx not in self.PPlabel.keys():
@@ -1411,6 +1472,7 @@ class MainWindow(QMainWindow, WindowMixin):
 
     def loadRecent(self, filename):
         if self.mayContinue():
+            print(filename,"======")
             self.loadFile(filename)
 
     def scanAllImages(self, folderPath):
@@ -1446,6 +1508,23 @@ class MainWindow(QMainWindow, WindowMixin):
         self.lastOpenDir = targetDirPath
         self.importDirImages(targetDirPath)
 
+    def openDatasetDirDialog(self,):
+        if self.lastOpenDir and os.path.exists(self.lastOpenDir):
+            if platform.system() == 'Windows':
+                os.startfile(self.lastOpenDir)
+            else:
+                os.system('open ' + os.path.normpath(self.lastOpenDir))
+            defaultOpenDirPath = self.lastOpenDir
+
+        else:
+            if self.lang == 'ch':
+                self.msgBox.warning(self, "提示", "\n 原文件夹已不存在,请从新选择数据集路径!")
+            else:
+                self.msgBox.warning(self, "Warn", "\n The original folder no longer exists, please choose the data set path again!")
+
+            self.actions.open_dataset_dir.setEnabled(False)
+            defaultOpenDirPath = os.path.dirname(self.filePath) if self.filePath else '.'
+
     def importDirImages(self, dirpath, isDelete = False):
         if not self.mayContinue() or not dirpath:
             return
@@ -1493,6 +1572,10 @@ class MainWindow(QMainWindow, WindowMixin):
         self.reRecogButton.setEnabled(True)
         self.actions.AutoRec.setEnabled(True)
         self.actions.reRec.setEnabled(True)
+        self.actions.open_dataset_dir.setEnabled(True)
+        self.actions.rotateLeft.setEnabled(True)
+        self.actions.rotateRight.setEnabled(True)
+
 
 
     def openPrevImg(self, _value=False):
@@ -1501,7 +1584,7 @@ class MainWindow(QMainWindow, WindowMixin):
 
         if self.filePath is None:
             return
-        
+
         currIndex = self.mImgList.index(self.filePath)
         self.mImgList5 = self.mImgList[:5]
         if currIndex - 1 >= 0:
@@ -1531,7 +1614,7 @@ class MainWindow(QMainWindow, WindowMixin):
         if filename:
             print('file name in openNext is ',filename)
             self.loadFile(filename)
-        
+
     def updateFileListIcon(self, filename):
         pass
 
@@ -1643,7 +1726,7 @@ class MainWindow(QMainWindow, WindowMixin):
         proc.startDetached(os.path.abspath(__file__))
 
     def mayContinue(self):  #
-        if not self.dirty:                                    
+        if not self.dirty:
             return True
         else:
             discardChanges = self.discardChangesDialog()
@@ -2037,6 +2120,8 @@ def read(filename, default=None):
     except:
         return default
 
+def str2bool(v):
+    return v.lower() in ("true", "t", "1")
 
 def get_main_app(argv=[]):
     """
@@ -2048,13 +2133,14 @@ def get_main_app(argv=[]):
     app.setWindowIcon(newIcon("app"))
     # Tzutalin 201705+: Accept extra agruments to change predefined class file
     argparser = argparse.ArgumentParser()
-    argparser.add_argument("--lang", default='en', nargs="?")
+    argparser.add_argument("--lang", type=str, default='en', nargs="?")
+    argparser.add_argument("--gpu", type=str2bool, default=False, nargs="?")
     argparser.add_argument("--predefined_classes_file",
                            default=os.path.join(os.path.dirname(__file__), "data", "predefined_classes.txt"),
                            nargs="?")
     args = argparser.parse_args(argv[1:])
     # Usage : labelImg.py image predefClassFile saveDir
-    win = MainWindow(lang=args.lang,
+    win = MainWindow(lang=args.lang, gpu=args.gpu,
                      defaultPrefdefClassFile=args.predefined_classes_file)
     win.show()
     return app, win
@@ -2067,7 +2153,7 @@ def main():
 
 
 if __name__ == '__main__':
-        
+
     resource_file = './libs/resources.py'
     if not os.path.exists(resource_file):
         output = os.system('pyrcc5 -o libs/resources.py resources.qrc')

PPOCRLabel/README.md

@@ -8,9 +8,12 @@ PPOCRLabel is a semi-automatic graphic annotation tool suitable for OCR field, w
 
 ### Recent Update
 
+- 2021.8.11：
+  - New functions: Open the dataset folder, image rotation (Note: Please delete the label box before rotating the image) (by [Wei-JL](https://github.com/Wei-JL))
+  - Added shortcut key description (Help-Shortcut Key), repaired the direction shortcut key movement function under batch processing (by [d2623587501](https://github.com/d2623587501))
 - 2021.2.5: New batch processing and undo functions (by [Evezerest](https://github.com/Evezerest)):
-  - Batch processing function: Press and hold the Ctrl key to select the box, you can move, copy, and delete in batches.
-  - Undo function: In the process of drawing a four-point label box or after editing the box, press Ctrl+Z to undo the previous operation.
+  - **Batch processing function**: Press and hold the Ctrl key to select the box, you can move, copy, and delete in batches.
+  - **Undo function**: In the process of drawing a four-point label box or after editing the box, press Ctrl+Z to undo the previous operation.
   - Fix image rotation and size problems, optimize the process of editing the mark frame (by [ninetailskim](https://github.com/ninetailskim)、 [edencfc](https://github.com/edencfc)).
 - 2021.1.11: Optimize the labeling experience (by [edencfc](https://github.com/edencfc)),
   - Users can choose whether to pop up the label input dialog after drawing the detection box in "View - Pop-up Label Input Dialog".
@@ -23,17 +26,51 @@ PPOCRLabel is a semi-automatic graphic annotation tool suitable for OCR field, w
 
 ## Installation
 
-### 1. Install PaddleOCR
+### 1. Environment Preparation
 
-PaddleOCR models has been built in PPOCRLabel, please refer to [PaddleOCR installation document](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/doc/doc_ch/installation.md) to prepare PaddleOCR and make sure it works.
+#### **Install PaddlePaddle 2.0**
 
-### 2. Install PPOCRLabel
+```bash
+pip3 install --upgrade pip
+
+# If you have cuda9 or cuda10 installed on your machine, please run the following command to install
+python3 -m pip install paddlepaddle-gpu -i https://mirror.baidu.com/pypi/simple
+
+# If you only have cpu on your machine, please run the following command to install
+python3 -m pip install paddlepaddle -i https://mirror.baidu.com/pypi/simple
+```
+
+For more software version requirements, please refer to the instructions in [Installation Document](https://www.paddlepaddle.org.cn/install/quick) for operation.
+
+#### **Install PaddleOCR**
+
+```bash
+# Recommend
+git clone https://github.com/PaddlePaddle/PaddleOCR
+
+# If you cannot pull successfully due to network problems, you can also choose to use the code hosting on the cloud:
+
+git clone https://gitee.com/paddlepaddle/PaddleOCR
 
-#### Windows + Anaconda
+# Note: The cloud-hosting code may not be able to synchronize the update with this GitHub project in real time. There might be a delay of 3-5 days. Please give priority to the recommended method.
+```
 
-Download and install [Anaconda](https://www.anaconda.com/download/#download) (Python 3+)
+#### **Install Third-party Libraries**
 
+```bash
+cd PaddleOCR
+pip3 install -r requirements.txt
 ```
+
+If you getting this error `OSError: [WinError 126] The specified module could not be found` when you install shapely on windows. Please try to download Shapely whl file using http://www.lfd.uci.edu/~gohlke/pythonlibs/#shapely.
+
+Reference: [Solve shapely installation on windows](https://stackoverflow.com/questions/44398265/install-shapely-oserror-winerror-126-the-specified-module-could-not-be-found)
+
+### 2. Install PPOCRLabel
+
+#### Windows
+
+```bash
 pip install pyqt5
 cd ./PPOCRLabel # Change the directory to the PPOCRLabel folder
 python PPOCRLabel.py
@@ -41,15 +78,15 @@ python PPOCRLabel.py
 
 #### Ubuntu Linux
 
-```
+```bash
 pip3 install pyqt5
 pip3 install trash-cli
 cd ./PPOCRLabel # Change the directory to the PPOCRLabel folder
 python3 PPOCRLabel.py
 ```
 
-#### macOS
-```
+#### MacOS
+```bash
 pip3 install pyqt5
 pip3 uninstall opencv-python # Uninstall opencv manually as it conflicts with pyqt
 pip3 install opencv-contrib-python-headless==4.2.0.32 # Install the headless version of opencv
@@ -79,11 +116,11 @@ python3 PPOCRLabel.py
 
 7. Double click the result in 'recognition result' list to manually change inaccurate recognition results.
 
-8. Click "Check", the image status will switch to "√",then the program automatically jump to the next.
+8. **Click "Check", the image status will switch to "√",then the program automatically jump to the next.**
 
 9. Click "Delete Image" and the image will be deleted to the recycle bin.
 
-10. Labeling result: the user can save manually through the menu "File - Save Label", while the program will also save automatically if "File - Auto Save Label Mode" is selected. The manually checked label will be stored in *Label.txt* under the opened picture folder. Click "PaddleOCR"-"Save Recognition Results" in the menu bar, the recognition training data of such pictures will be saved in the *crop_img* folder, and the recognition label will be saved in *rec_gt.txt*<sup>[4]</sup>.
+10. Labeling result: the user can export the label result manually through the menu "File - Export Label", while the program will also export automatically if "File - Auto export Label Mode" is selected. The manually checked label will be stored in *Label.txt* under the opened picture folder. Click "File"-"Export Recognition Results" in the menu bar, the recognition training data of such pictures will be saved in the *crop_img* folder, and the recognition label will be saved in *rec_gt.txt*<sup>[4]</sup>.
 
 ### Note
 
@@ -97,10 +134,10 @@ python3 PPOCRLabel.py
 
 |   File name   |                         Description                          |
 | :-----------: | :----------------------------------------------------------: |
-|   Label.txt   | The detection label file can be directly used for PPOCR detection model training. After the user saves 5 label results, the file will be automatically saved. It will also be written when the user closes the application or changes the file folder. |
+|   Label.txt   | The detection label file can be directly used for PPOCR detection model training. After the user saves 5 label results, the file will be automatically exported. It will also be written when the user closes the application or changes the file folder. |
 | fileState.txt | The picture status file save the image in the current folder that has been manually confirmed by the user. |
 |  Cache.cach   |    Cache files to save the results of model recognition.     |
-|  rec_gt.txt   | The recognition label file, which can be directly used for PPOCR identification model training, is generated after the user clicks on the menu bar "File"-"Save recognition result". |
+|  rec_gt.txt   | The recognition label file, which can be directly used for PPOCR identification model training, is generated after the user clicks on the menu bar "File"-"Export recognition result". |
 |   crop_img    | The recognition data, generated at the same time with *rec_gt.txt* |
 
 ## Explanation
@@ -134,16 +171,16 @@ python3 PPOCRLabel.py
 
 - Custom model: The model trained by users can be replaced by modifying PPOCRLabel.py in [PaddleOCR class instantiation](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/PPOCRLabel/PPOCRLabel.py#L110) referring [Custom Model Code](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/doc/doc_en/whl_en.md#use-custom-model)
 
-### Save
+### Export Label Result
 
-PPOCRLabel supports three ways to save Label.txt
+PPOCRLabel supports three ways to export Label.txt
 
-- Automatically save: After selecting "File - Auto Save Label Mode", the program will automatically write the annotations into Label.txt every time the user confirms an image. If this option is not turned on, it will be automatically saved after detecting that the user has manually checked 5 images.
-- Manual save: Click "File-Save Marking Results" to manually save the label.
-- Close application save
+- Automatically export: After selecting "File - Auto Export Label Mode", the program will automatically write the annotations into Label.txt every time the user confirms an image. If this option is not turned on, it will be automatically exported after detecting that the user has manually checked 5 images.
+- Manual export: Click "File-Export Marking Results" to manually export the label.
+- Close application export
 
 
-### Export partial recognition results
+### Export Partial Recognition Results
 
 For some data that are difficult to recognize, the recognition results will not be exported by **unchecking** the corresponding tags in the recognition results checkbox.
 

PPOCRLabel/README_ch.md

@@ -8,9 +8,12 @@ PPOCRLabel是一款适用于OCR领域的半自动化图形标注工具，内置P
 
 #### 近期更新
 
+- 2021.8.11：
+  - 新增功能：打开数据所在文件夹、图像旋转（注意：旋转前的图片上不能存在标记框）（by [Wei-JL](https://github.com/Wei-JL)）
+  - 新增快捷键说明（帮助-快捷键）、修复批处理下的方向快捷键移动功能（by [d2623587501](https://github.com/d2623587501)）
 - 2021.2.5：新增批处理与撤销功能（by [Evezerest](https://github.com/Evezerest))
-  - 批处理功能：按住Ctrl键选择标记框后可批量移动、复制、删除。
-  - 撤销功能：在绘制四点标注框过程中或对框进行编辑操作后，按下Ctrl+Z可撤销上一部操作。
+  - **批处理功能**：按住Ctrl键选择标记框后可批量移动、复制、删除、重新识别。
+  - **撤销功能**：在绘制四点标注框过程中或对框进行编辑操作后，按下Ctrl+Z可撤销上一部操作。
   - 修复图像旋转和尺寸问题、优化编辑标记框过程（by [ninetailskim](https://github.com/ninetailskim)、 [edencfc](https://github.com/edencfc)）
 - 2021.1.11：优化标注体验（by [edencfc](https://github.com/edencfc)）：
   - 用户可在“视图 - 弹出标记输入框”选择在画完检测框后标记输入框是否弹出。
@@ -27,13 +30,48 @@ PPOCRLabel是一款适用于OCR领域的半自动化图形标注工具，内置P
 
 ## 安装
 
-### 1. 安装PaddleOCR
-PPOCRLabel内置PaddleOCR模型，故请参考[PaddleOCR安装文档](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/doc/doc_ch/installation.md)准备好PaddleOCR，并确保PaddleOCR安装成功。
+### 1. 环境搭建
+#### 安装PaddlePaddle
 
-### 2. 安装PPOCRLabel
-#### Windows + Anaconda
+```bash
+pip3 install --upgrade pip
+
+如果您的机器安装的是CUDA9或CUDA10，请运行以下命令安装
+python3 -m pip install paddlepaddle-gpu -i https://mirror.baidu.com/pypi/simple
+
+如果您的机器是CPU，请运行以下命令安装
+
+python3 -m pip install paddlepaddle -i https://mirror.baidu.com/pypi/simple
+```
+
+更多的版本需求，请参照[安装文档](https://www.paddlepaddle.org.cn/install/quick)中的说明进行操作。
+
+#### **安装PaddleOCR**
+
+```bash
+【推荐】git clone https://github.com/PaddlePaddle/PaddleOCR
+
+如果因为网络问题无法pull成功，也可选择使用码云上的托管：
+
+git clone https://gitee.com/paddlepaddle/PaddleOCR
+
+注：码云托管代码可能无法实时同步本github项目更新，存在3~5天延时，请优先使用推荐方式。
+```
+
+#### 安装第三方库
 
+```bash
+cd PaddleOCR
+pip3 install -r requirements.txt
 ```
+
+注意，windows环境下，建议从[这里](https://www.lfd.uci.edu/~gohlke/pythonlibs/#shapely)下载shapely安装包完成安装， 直接通过pip安装的shapely库可能出现`[winRrror 126] 找不到指定模块的问题`。
+
+### 2. 安装PPOCRLabel
+
+#### Windows
+
+```bash
 pip install pyqt5
 cd ./PPOCRLabel # 将目录切换到PPOCRLabel文件夹下
 python PPOCRLabel.py --lang ch
@@ -41,15 +79,15 @@ python PPOCRLabel.py --lang ch
 
 #### Ubuntu Linux
 
-```
+```bash
 pip3 install pyqt5
 pip3 install trash-cli
 cd ./PPOCRLabel # 将目录切换到PPOCRLabel文件夹下
 python3 PPOCRLabel.py --lang ch
 ```
 
-#### macOS
-```
+#### MacOS
+```bash
 pip3 install pyqt5
 pip3 uninstall opencv-python # 由于mac版本的opencv与pyqt有冲突，需先手动卸载opencv
 pip3 install opencv-contrib-python-headless==4.2.0.32 # 安装headless版本的open-cv
@@ -57,6 +95,8 @@ cd ./PPOCRLabel # 将目录切换到PPOCRLabel文件夹下
 python3 PPOCRLabel.py --lang ch
 ```
 
+
+
 ## 使用
 
 ### 操作步骤
@@ -68,9 +108,9 @@ python3 PPOCRLabel.py --lang ch
 5. 标记框绘制完成后，用户点击 “确认”，检测框会先被预分配一个 “待识别” 标签。
 6. 重新识别：将图片中的所有检测画绘制/调整完成后，点击 “重新识别”，PPOCR模型会对当前图片中的**所有检测框**重新识别<sup>[3]</sup>。
 7. 内容更改：双击识别结果，对不准确的识别结果进行手动更改。
-8. **确认标记**：点击 “确认”，图片状态切换为 “√”，跳转至下一张。
+8. **确认标记：点击 “确认”，图片状态切换为 “√”，跳转至下一张。**
 9. 删除：点击 “删除图像”，图片将会被删除至回收站。
-10. 保存结果：用户可以通过菜单中“文件-保存标记结果”手动保存，同时也可以点击“文件 - 自动保存标记结果”开启自动保存。手动确认过的标记将会被存放在所打开图片文件夹下的*Label.txt*中。在菜单栏点击 “文件” - "保存识别结果"后，会将此类图片的识别训练数据保存在*crop_img*文件夹下，识别标签保存在*rec_gt.txt*中<sup>[4]</sup>。
+10. 导出结果：用户可以通过菜单中“文件-导出标记结果”手动导出，同时也可以点击“文件 - 自动导出标记结果”开启自动导出。手动确认过的标记将会被存放在所打开图片文件夹下的*Label.txt*中。在菜单栏点击 “文件” - "导出识别结果"后，会将此类图片的识别训练数据保存在*crop_img*文件夹下，识别标签保存在*rec_gt.txt*中<sup>[4]</sup>。
 
 ### 注意
 
@@ -84,10 +124,10 @@ python3 PPOCRLabel.py --lang ch
 
 |    文件名     |                             说明                             |
 | :-----------: | :----------------------------------------------------------: |
-|   Label.txt   | 检测标签，可直接用于PPOCR检测模型训练。用户每保存5张检测结果后，程序会进行自动写入。当用户关闭应用程序或切换文件路径后同样会进行写入。 |
+|   Label.txt   | 检测标签，可直接用于PPOCR检测模型训练。用户每确认5张检测结果后，程序会进行自动写入。当用户关闭应用程序或切换文件路径后同样会进行写入。 |
 | fileState.txt | 图片状态标记文件，保存当前文件夹下已经被用户手动确认过的图片名称。 |
 |  Cache.cach   |              缓存文件，保存模型自动识别的结果。              |
-|  rec_gt.txt   | 识别标签。可直接用于PPOCR识别模型训练。需用户手动点击菜单栏“文件” - "保存识别结果"后产生。 |
+|  rec_gt.txt   | 识别标签。可直接用于PPOCR识别模型训练。需用户手动点击菜单栏“文件” - "导出识别结果"后产生。 |
 |   crop_img    |   识别数据。按照检测框切割后的图片。与rec_gt.txt同时产生。   |
 
 ## 说明
@@ -120,19 +160,19 @@ python3 PPOCRLabel.py --lang ch
 
  - 自定义模型：用户可根据[自定义模型代码使用](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/doc/doc_ch/whl.md#%E8%87%AA%E5%AE%9A%E4%B9%89%E6%A8%A1%E5%9E%8B)，通过修改PPOCRLabel.py中针对[PaddleOCR类的实例化](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/PPOCRLabel/PPOCRLabel.py#L110)替换成自己训练的模型。
 
-### 保存方式
+### 导出标记结果
 
-PPOCRLabel支持三种保存方式：
+PPOCRLabel支持三种导出方式：
 
-- 自动保存：点击“文件 - 自动保存标记结果”后，用户每确认过一张图片，程序自动将标记结果写入Label.txt中。若未开启此选项，则检测到用户手动确认过5张图片后进行自动保存。
-- 手动保存：点击“文件 - 保存标记结果”手动保存标记。
-- 关闭应用程序保存
+- 自动导出：点击“文件 - 自动导出标记结果”后，用户每确认过一张图片，程序自动将标记结果写入Label.txt中。若未开启此选项，则检测到用户手动确认过5张图片后进行自动导出。
+- 手动导出：点击“文件 - 导出标记结果”手动导出标记。
+- 关闭应用程序导出
 
 ### 导出部分识别结果
 
 针对部分难以识别的数据，通过在识别结果的复选框中**取消勾选**相应的标记，其识别结果不会被导出。
 
-*注意：识别结果中的复选框状态仍需用户手动点击保存后才能保留*
+*注意：识别结果中的复选框状态仍需用户手动点击确认后才能保留*
 
 ### 错误提示
 - 如果同时使用whl包安装了paddleocr，其优先级大于通过paddleocr.py调用PaddleOCR类，whl包未更新时会导致程序异常。

PPOCRLabel/libs/canvas.py

@@ -23,6 +23,7 @@ except ImportError:
 
 from libs.shape import Shape
 from libs.utils import distance
+import copy
 
 CURSOR_DEFAULT = Qt.ArrowCursor
 CURSOR_POINT = Qt.PointingHandCursor
@@ -81,6 +82,7 @@ class Canvas(QWidget):
         self.fourpoint = True # ADD
         self.pointnum = 0
         self.movingShape = False
+        self.selectCountShape = False
 
         #initialisation for panning
         self.pan_initial_pos = QPoint()
@@ -702,6 +704,10 @@ class Canvas(QWidget):
 
     def keyPressEvent(self, ev):
         key = ev.key()
+        shapesBackup = []
+        shapesBackup = copy.deepcopy(self.shapes)
+        self.shapesBackups.pop()
+        self.shapesBackups.append(shapesBackup)
         if key == Qt.Key_Escape and self.current:
             print('ESC press')
             self.current = None
@@ -709,41 +715,48 @@ class Canvas(QWidget):
             self.update()
         elif key == Qt.Key_Return and self.canCloseShape():
             self.finalise()
-        elif key == Qt.Key_Left and self.selectedShape:
+        elif key == Qt.Key_Left and self.selectedShapes:
              self.moveOnePixel('Left')
-        elif key == Qt.Key_Right and self.selectedShape:
+        elif key == Qt.Key_Right and self.selectedShapes:
              self.moveOnePixel('Right')
-        elif key == Qt.Key_Up and self.selectedShape:
+        elif key == Qt.Key_Up and self.selectedShapes:
              self.moveOnePixel('Up')
-        elif key == Qt.Key_Down and self.selectedShape:
+        elif key == Qt.Key_Down and self.selectedShapes:
              self.moveOnePixel('Down')
 
     def moveOnePixel(self, direction):
         # print(self.selectedShape.points)
-        if direction == 'Left' and not self.moveOutOfBound(QPointF(-1.0, 0)):
-            # print("move Left one pixel")
-            self.selectedShape.points[0] += QPointF(-1.0, 0)
-            self.selectedShape.points[1] += QPointF(-1.0, 0)
-            self.selectedShape.points[2] += QPointF(-1.0, 0)
-            self.selectedShape.points[3] += QPointF(-1.0, 0)
-        elif direction == 'Right' and not self.moveOutOfBound(QPointF(1.0, 0)):
-            # print("move Right one pixel")
-            self.selectedShape.points[0] += QPointF(1.0, 0)
-            self.selectedShape.points[1] += QPointF(1.0, 0)
-            self.selectedShape.points[2] += QPointF(1.0, 0)
-            self.selectedShape.points[3] += QPointF(1.0, 0)
-        elif direction == 'Up' and not self.moveOutOfBound(QPointF(0, -1.0)):
-            # print("move Up one pixel")
-            self.selectedShape.points[0] += QPointF(0, -1.0)
-            self.selectedShape.points[1] += QPointF(0, -1.0)
-            self.selectedShape.points[2] += QPointF(0, -1.0)
-            self.selectedShape.points[3] += QPointF(0, -1.0)
-        elif direction == 'Down' and not self.moveOutOfBound(QPointF(0, 1.0)):
-            # print("move Down one pixel")
-            self.selectedShape.points[0] += QPointF(0, 1.0)
-            self.selectedShape.points[1] += QPointF(0, 1.0)
-            self.selectedShape.points[2] += QPointF(0, 1.0)
-            self.selectedShape.points[3] += QPointF(0, 1.0)
+        self.selectCount = len(self.selectedShapes)
+        self.selectCountShape = True
+        for i in range(len(self.selectedShapes)):
+            self.selectedShape = self.selectedShapes[i]
+            if direction == 'Left' and not self.moveOutOfBound(QPointF(-1.0, 0)):
+                # print("move Left one pixel")
+                self.selectedShape.points[0] += QPointF(-1.0, 0)
+                self.selectedShape.points[1] += QPointF(-1.0, 0)
+                self.selectedShape.points[2] += QPointF(-1.0, 0)
+                self.selectedShape.points[3] += QPointF(-1.0, 0)
+            elif direction == 'Right' and not self.moveOutOfBound(QPointF(1.0, 0)):
+                # print("move Right one pixel")
+                self.selectedShape.points[0] += QPointF(1.0, 0)
+                self.selectedShape.points[1] += QPointF(1.0, 0)
+                self.selectedShape.points[2] += QPointF(1.0, 0)
+                self.selectedShape.points[3] += QPointF(1.0, 0)
+            elif direction == 'Up' and not self.moveOutOfBound(QPointF(0, -1.0)):
+                # print("move Up one pixel")
+                self.selectedShape.points[0] += QPointF(0, -1.0)
+                self.selectedShape.points[1] += QPointF(0, -1.0)
+                self.selectedShape.points[2] += QPointF(0, -1.0)
+                self.selectedShape.points[3] += QPointF(0, -1.0)
+            elif direction == 'Down' and not self.moveOutOfBound(QPointF(0, 1.0)):
+                # print("move Down one pixel")
+                self.selectedShape.points[0] += QPointF(0, 1.0)
+                self.selectedShape.points[1] += QPointF(0, 1.0)
+                self.selectedShape.points[2] += QPointF(0, 1.0)
+                self.selectedShape.points[3] += QPointF(0, 1.0)
+        shapesBackup = []
+        shapesBackup = copy.deepcopy(self.shapes)
+        self.shapesBackups.append(shapesBackup)
         self.shapeMoved.emit()
         self.repaint()
 
@@ -840,6 +853,7 @@ class Canvas(QWidget):
     def restoreShape(self):
         if not self.isShapeRestorable:
             return
+
         self.shapesBackups.pop()  # latest
         shapesBackup = self.shapesBackups.pop()
         self.shapes = shapesBackup

PPOCRLabel/libs/utils.py

@@ -124,6 +124,15 @@ def natural_sort(list, key=lambda s:s):
 
 
 def get_rotate_crop_image(img, points):
+    # Use Green's theory to judge clockwise or counterclockwise
+    # author: biyanhua
+    d = 0.0
+    for index in range(-1, 3):
+        d += -0.5 * (points[index + 1][1] + points[index][1]) * (
+                    points[index + 1][0] - points[index][0])
+    if d < 0: # counterclockwise
+        tmp = np.array(points)
+        points[1], points[3] = tmp[3], tmp[1]
 
     try:
         img_crop_width = int(
@@ -165,6 +174,7 @@ def stepsInfo(lang='en'):
               "10. 标注结果：关闭应用程序或切换文件路径后，手动保存过的标签将会被存放在所打开图片文件夹下的" \
               "*Label.txt*中。在菜单栏点击 “PaddleOCR” - 保存识别结果后，会将此类图片的识别训练数据保存在*crop_img*文件夹下，" \
               "识别标签保存在*rec_gt.txt*中。\n"
+
     else:
         msg = "1. Build and launch using the instructions above.\n" \
               "2. Click 'Open Dir' in Menu/File to select the folder of the picture.\n"\
@@ -178,5 +188,57 @@ def stepsInfo(lang='en'):
               "8. Click 'Save', the image status will switch to '√',then the program automatically jump to the next.\n"\
               "9. Click 'Delete Image' and the image will be deleted to the recycle bin.\n"\
               "10. Labeling result: After closing the application or switching the file path, the manually saved label will be stored in *Label.txt* under the opened picture folder.\n"\
-              "    Click PaddleOCR-Save Recognition Results in the menu bar, the recognition training data of such pictures will be saved in the *crop_img* folder, and the recognition label will be saved in *rec_gt.txt*.\n"
+              "    Click PaddleOCR-Save Recognition Results in the menu bar, the recognition training data of such pictures will be saved in the *crop_img* folder, and the recognition label will be saved in *rec_gt.txt*.\n" 
+
+    return msg
+
+def keysInfo(lang='en'):
+    if lang == 'ch':
+        msg = "快捷键\t\t\t说明\n" \
+              "———————————————————————\n"\
+              "Ctrl + shift + R\t\t对当前图片的所有标记重新识别\n" \
+              "W\t\t\t新建矩形框\n" \
+              "Q\t\t\t新建四点框\n" \
+              "Ctrl + E\t\t编辑所选框标签\n" \
+              "Ctrl + R\t\t重新识别所选标记\n" \
+              "Ctrl + C\t\t复制并粘贴选中的标记框\n" \
+              "Ctrl + 鼠标左键\t\t多选标记框\n" \
+              "Backspace\t\t删除所选框\n" \
+              "Ctrl + V\t\t确认本张图片标记\n" \
+              "Ctrl + Shift + d\t删除本张图片\n" \
+              "D\t\t\t下一张图片\n" \
+              "A\t\t\t上一张图片\n" \
+              "Ctrl++\t\t\t缩小\n" \
+              "Ctrl--\t\t\t放大\n" \
+              "↑→↓←\t\t\t移动标记框\n" \
+              "———————————————————————\n" \
+              "注：Mac用户Command键替换上述Ctrl键"
+
+    else:
+        msg = "Shortcut Keys\t\tDescription\n" \
+              "———————————————————————\n" \
+              "Ctrl + shift + R\t\tRe-recognize all the labels\n" \
+              "\t\t\tof the current image\n" \
+              "\n"\
+              "W\t\t\tCreate a rect box\n" \
+              "Q\t\t\tCreate a four-points box\n" \
+              "Ctrl + E\t\tEdit label of the selected box\n" \
+              "Ctrl + R\t\tRe-recognize the selected box\n" \
+              "Ctrl + C\t\tCopy and paste the selected\n" \
+              "\t\t\tbox\n" \
+              "\n"\
+              "Ctrl + Left Mouse\tMulti select the label\n" \
+              "Button\t\t\tbox\n" \
+              "\n"\
+              "Backspace\t\tDelete the selected box\n" \
+              "Ctrl + V\t\tCheck image\n" \
+              "Ctrl + Shift + d\tDelete image\n" \
+              "D\t\t\tNext image\n" \
+              "A\t\t\tPrevious image\n" \
+              "Ctrl++\t\t\tZoom in\n" \
+              "Ctrl--\t\t\tZoom out\n" \
+              "↑→↓←\t\t\tMove selected box" \
+              "———————————————————————\n" \
+              "Notice:For Mac users, use the 'Command' key instead of the 'Ctrl' key"
+
     return msg
\ No newline at end of file

PPOCRLabel/resources.qrc

@@ -18,6 +18,8 @@
 <file alias="quit">resources/icons/quit.png</file>
 <file alias="copy">resources/icons/copy.png</file>
 <file alias="edit">resources/icons/edit.png</file>
+<file alias="rotateLeft">resources/icons/rotateLeft.png</file>
+<file alias="rotateRight">resources/icons/rotateRight.png</file>
 <file alias="open">resources/icons/open.png</file>
 <file alias="save">resources/icons/save.png</file>
 <file alias="format_voc">resources/icons/format_voc.png</file>

PPOCRLabel/resources/strings/strings-zh-CN.properties

@@ -31,6 +31,7 @@ save=确认
 saveAs=另存为
 fitWinDetail=缩放到当前窗口大小
 openDir=打开目录
+openDatasetDir=打开数据集路径
 copyPrevBounding=复制当前图像中的上一个边界框
 showHide=显示/隐藏标签
 changeSaveFormat=更改存储格式
@@ -85,19 +86,22 @@ detectionBoxposition=检测框位置
 recognitionResult=识别结果
 creatPolygon=四点标注
 drawSquares=正方形标注
-saveRec=保存识别结果
+rotateLeft=图片左旋转90度
+rotateRight=图片右旋转90度
+saveRec=导出识别结果
 tempLabel=待识别
 nullLabel=无法识别
 steps=操作步骤
+keys=快捷键
 choseModelLg=选择模型语言
 cancel=取消
 ok=确认
 autolabeling=自动标注中
 hideBox=隐藏所有标注
 showBox=显示所有标注
-saveLabel=保存标记结果
+saveLabel=导出标记结果
 singleRe=重识别此区块
 labelDialogOption=弹出标记输入框
 undo=撤销
 undoLastPoint=撤销上个点
-autoSaveMode=自动保存标记结果
\ No newline at end of file
+autoSaveMode=自动导出标记结果
\ No newline at end of file

PPOCRLabel/resources/strings/strings.properties

@@ -3,6 +3,7 @@ openFileDetail=Open image or label file
 quit=Quit
 quitApp=Quit application
 openDir=Open Dir
+openDatasetDir=Open DatasetDir
 copyPrevBounding=Copy previous Bounding Boxes in the current image 
 changeSavedAnnotationDir=Change default saved Annotation dir
 openAnnotation=Open Annotation
@@ -84,20 +85,23 @@ iconList=Icon List
 detectionBoxposition=Detection box position
 recognitionResult=Recognition result
 creatPolygon=Create Quadrilateral
+rotateLeft=Left turn 90 degrees
+rotateRight=Right turn 90 degrees
 drawSquares=Draw Squares
-saveRec=Save Recognition Result
+saveRec=Export Recognition Result
 tempLabel=TEMPORARY
 nullLabel=NULL
 steps=Steps
+keys=Shortcut Keys
 choseModelLg=Choose Model Language
 cancel=Cancel
 ok=OK
 autolabeling=Automatic Labeling
 hideBox=Hide All Box
 showBox=Show All Box
-saveLabel=Save Label
+saveLabel=Export Label
 singleRe=Re-recognition RectBox
 labelDialogOption=Pop-up Label Input Dialog
 undo=Undo
 undoLastPoint=Undo Last Point
-autoSaveMode=Auto Save Label Mode
\ No newline at end of file
+autoSaveMode=Auto Export Label Mode
\ No newline at end of file

README.md

 English | [简体中文](README_ch.md)
 
+<p align="center">
+ <img src="./doc/PaddleOCR_log.png" align="middle" width = "600"/>
+<p align="center">
+
+
+------------------------------------------------------------------------------------------
+
+<p align="left">
+    <a href="./LICENSE"><img src="https://img.shields.io/badge/license-Apache%202-dfd.svg"></a>
+    <a href="https://github.com/PaddlePaddle/PaddleOCR/releases"><img src="https://img.shields.io/github/v/release/PaddlePaddle/PaddleOCR?color=ffa"></a>
+    <a href=""><img src="https://img.shields.io/badge/python-3.7+-aff.svg"></a>
+    <a href=""><img src="https://img.shields.io/badge/os-linux%2C%20win%2C%20mac-pink.svg"></a>
+    <a href=""><img src="https://img.shields.io/pypi/format/PaddleOCR?color=c77"></a>
+    <a href="https://github.com/PaddlePaddle/PaddleOCR/graphs/contributors"><img src="https://img.shields.io/github/contributors/PaddlePaddle/PaddleOCR?color=9ea"></a>
+    <a href="https://pypi.org/project/PaddleOCR/"><img src="https://img.shields.io/pypi/dm/PaddleOCR?color=9cf"></a>
+    <a href="https://github.com/PaddlePaddle/PaddleOCR/stargazers"><img src="https://img.shields.io/github/stars/PaddlePaddle/PaddleOCR?color=ccf"></a>
+</p>
+
 ## Introduction
+
 PaddleOCR aims to create multilingual, awesome, leading, and practical OCR tools that help users train better models and apply them into practice.
 
-## Notice
-PaddleOCR supports both dynamic graph and static graph programming paradigm
-- Dynamic graph: dygraph branch (default), **supported by paddle 2.0.0 ([installation](./doc/doc_en/installation_en.md))**
-- Static graph: develop branch
 
 **Recent updates**
-- 2021.1.21 update more than 25+ multilingual recognition models [models list](./doc/doc_en/models_list_en.md), including：English, Chinese, German, French, Japanese，Spanish，Portuguese Russia Arabic and so on.  Models for more languages will continue to be updated [Develop Plan](https://github.com/PaddlePaddle/PaddleOCR/issues/1048).
-- 2020.12.15 update Data synthesis tool, i.e., [Style-Text](./StyleText/README.md)，easy to synthesize a large number of images which are similar to the target scene image.
-- 2020.11.25 Update a new data annotation tool, i.e., [PPOCRLabel](./PPOCRLabel/README.md), which is helpful to improve the labeling efficiency. Moreover, the labeling results can be used in training of the PP-OCR system directly.
-- 2020.9.22 Update the PP-OCR technical article, https://arxiv.org/abs/2009.09941
+
+- PaddleOCR R&D team would like to share the key points of PP-OCRv2, at 20:15 pm on September 8th, [Live Address](https://live.bilibili.com/21689802).
+- 2021.9.7 release PaddleOCR v2.3, [PP-OCRv2](#PP-OCRv2) is proposed. The inference speed of PP-OCRv2 is 220% higher than that of PP-OCR server in CPU device. The F-score of PP-OCRv2 is 7% higher than that of PP-OCR mobile.
+- 2021.8.3 released PaddleOCR v2.2, add a new structured documents analysis toolkit, i.e., [PP-Structure](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.2/ppstructure/README.md), support layout analysis and table recognition (One-key to export chart images to Excel files).
+- 2021.4.8 release end-to-end text recognition algorithm [PGNet](https://www.aaai.org/AAAI21Papers/AAAI-2885.WangP.pdf) which is published in AAAI 2021. Find tutorial [here](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.1/doc/doc_en/pgnet_en.md)；release multi language recognition [models](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.1/doc/doc_en/multi_languages_en.md), support more than 80 languages recognition; especically, the performance of [English recognition model](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.1/doc/doc_en/models_list_en.md#English) is Optimized.
+
 - [more](./doc/doc_en/update_en.md)
 
 ## Features
-- PPOCR series of high-quality pre-trained models, comparable to commercial effects
-    - Ultra lightweight ppocr_mobile series models: detection (3.0M) + direction classifier (1.4M) + recognition (5.0M) = 9.4M
-    - General ppocr_server series models: detection (47.1M) + direction classifier (1.4M) + recognition (94.9M) = 143.4M
+- PP-OCR series of high-quality pre-trained models, comparable to commercial effects
+    - Ultra lightweight PP-OCRv2 series models: detection (3.1M) + direction classifier (1.4M) + recognition 8.5M) = 13.0M
+    - Ultra lightweight PP-OCR mobile series models: detection (3.0M) + direction classifier (1.4M) + recognition (5.0M) = 9.4M
+    - General PP-OCR server series models: detection (47.1M) + direction classifier (1.4M) + recognition (94.9M) = 143.4M
     - Support Chinese, English, and digit recognition, vertical text recognition, and long text recognition
     - Support multi-language recognition: Korean, Japanese, German, French
 - Rich toolkits related to the OCR areas
@@ -43,7 +61,7 @@ The above pictures are the visualizations of the general ppocr_server model. For
 - Scan the QR code below with your Wechat, you can access to official technical exchange group. Look forward to your participation.
 
 <div align="center">
-<img src="https://raw.githubusercontent.com/PaddlePaddle/PaddleOCR/release/2.0/doc/joinus.PNG"  width = "200" height = "200" />
+<img src="https://raw.githubusercontent.com/PaddlePaddle/PaddleOCR/dygraph/doc/joinus.PNG"  width = "200" height = "200" />
 </div>
 
 
@@ -64,39 +82,45 @@ Mobile DEMO experience (based on EasyEdge and Paddle-Lite, supports iOS and Andr
 <a name="Supported-Chinese-model-list"></a>
 
 
-## PP-OCR 2.0 series model list（Update on Dec 15）
-**Note** : Compared with [models 1.1](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/doc/doc_en/models_list_en.md), which are trained with static graph programming paradigm, models 2.0 are the dynamic graph trained version and achieve close performance.
+## PP-OCR Series Model List（Update on September 8th）
 
 | Model introduction                                           | Model name                   | Recommended scene | Detection model                                              | Direction classifier                                         | Recognition model                                            |
 | ------------------------------------------------------------ | ---------------------------- | ----------------- | ------------------------------------------------------------ | ------------------------------------------------------------ | ------------------------------------------------------------ |
-| Chinese and English ultra-lightweight OCR model (9.4M)       | ch_ppocr_mobile_v2.0_xx      | Mobile & server   |[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_train.tar)|[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_train.tar) |[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_pre.tar)      |
-| Chinese and English general OCR model (143.4M)               | ch_ppocr_server_v2.0_xx      | Server            |[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_infer.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_train.tar)    |[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_traingit.tar)    |[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_infer.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_pre.tar)  |  
+| Chinese and English ultra-lightweight PP-OCRv2 model（11.6M） |  ch_PP-OCRv2_xx |Mobile&Server|[inference model](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_det_infer.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_det_distill_train.tar)| [inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_train.tar) |[inference model](https://paddleocr.bj.bcebos.com/PP-OCRv2/ch/ch_PP-OCRv2_rec_infer.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_train.tar)|
+| Chinese and English ultra-lightweight PP-OCR model (9.4M)       | ch_ppocr_mobile_v2.0_xx      | Mobile & server   |[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_train.tar)|[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_train.tar) |[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_pre.tar)      |
+| Chinese and English general PP-OCR model (143.4M)               | ch_ppocr_server_v2.0_xx      | Server            |[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_infer.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_train.tar)    |[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_traingit.tar)    |[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_infer.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_pre.tar)  |  
 
 
-For more model downloads (including multiple languages), please refer to [PP-OCR v2.0 series model downloads](./doc/doc_en/models_list_en.md).
+For more model downloads (including multiple languages), please refer to [PP-OCR series model downloads](./doc/doc_en/models_list_en.md).
 
 For a new language request, please refer to [Guideline for new language_requests](#language_requests).
 
 ## Tutorials
-- [Installation](./doc/doc_en/installation_en.md)
+- [Environment Preparation](./doc/doc_en/environment_en.md)
 - [Quick Start](./doc/doc_en/quickstart_en.md)
-- [Code Structure](./doc/doc_en/tree_en.md)
-- Algorithm Introduction
-    - [Text Detection Algorithm](./doc/doc_en/algorithm_overview_en.md)
-    - [Text Recognition Algorithm](./doc/doc_en/algorithm_overview_en.md)
-    - [PP-OCR Pipeline](#PP-OCR-Pipeline)
-- Model Training/Evaluation
-    - [Text Detection](./doc/doc_en/detection_en.md)
-    - [Text Recognition](./doc/doc_en/recognition_en.md)
-    - [Direction Classification](./doc/doc_en/angle_class_en.md)
-    - [Yml Configuration](./doc/doc_en/config_en.md)
-- Inference and Deployment
-    - [Quick Inference Based on PIP](./doc/doc_en/whl_en.md)
+- [PaddleOCR Overview and Installation](./doc/doc_en/paddleOCR_overview_en.md)
+- PP-OCR Industry Landing: from Training to Deployment
+    - [PP-OCR Model and Configuration](./doc/doc_en/models_and_config_en.md)
+        - [PP-OCR Model Download](./doc/doc_en/models_list_en.md)
+        - [Yml Configuration](./doc/doc_en/config_en.md)
+        - [Python Inference for PP-OCR Model Library](./doc/doc_en/inference_ppocr_en.md)
+    - [PP-OCR Training](./doc/doc_en/training_en.md)
+        - [Text Detection](./doc/doc_en/detection_en.md)
+        - [Text Recognition](./doc/doc_en/recognition_en.md)
+        - [Text Direction Classification](./doc/doc_en/angle_class_en.md)
+        - [Yml Configuration](./doc/doc_en/config_en.md)
+    - Inference and Deployment
+        - [C++ Inference](./deploy/cpp_infer/readme_en.md)
+        - [Serving](./deploy/pdserving/README.md)
+        - [Mobile](./deploy/lite/readme_en.md)
+        - [Benchmark](./doc/doc_en/benchmark_en.md)  
+- [PP-Structure: Information Extraction](./ppstructure/README.md)
+    - [Layout Parser](./ppstructure/layout/README.md)
+    - [Table Recognition](./ppstructure/table/README.md)
+- Academic Circles
+    - [Two-stage Algorithm](./doc/doc_en/algorithm_overview_en.md)
+    - [PGNet Algorithm](./doc/doc_en/algorithm_overview_en.md)
     - [Python Inference](./doc/doc_en/inference_en.md)
-    - [C++ Inference](./deploy/cpp_infer/readme_en.md)
-    - [Serving](./deploy/pdserving/README.md)
-    - [Mobile](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/deploy/lite/readme_en.md)
-    - [Benchmark](./doc/doc_en/benchmark_en.md)  
 - Data Annotation and Synthesis
     - [Semi-automatic Annotation Tool: PPOCRLabel](./PPOCRLabel/README.md)
     - [Data Synthesis Tool: Style-Text](./StyleText/README.md)
@@ -114,17 +138,18 @@ For a new language request, please refer to [Guideline for new language_requests
 - [License](#LICENSE)
 - [Contribution](#CONTRIBUTION)
 
+<a name="PP-OCRv2"></a>
 
+## PP-OCRv2 Pipeline
+<div align="center">
+    <img src="./doc/ppocrv2_framework.jpg" width="800">
+</div>
 
-<a name="PP-OCR-Pipeline"></a>
+[1] PP-OCR is a practical ultra-lightweight OCR system. It is mainly composed of three parts: DB text detection, detection frame correction and CRNN text recognition. The system adopts 19 effective strategies from 8 aspects including backbone network selection and adjustment, prediction head design, data augmentation, learning rate transformation strategy, regularization parameter selection, pre-training model use, and automatic model tailoring and quantization to optimize and slim down the models of each module (as shown in the green box above). The final results are an ultra-lightweight Chinese and English OCR model with an overall size of 3.5M and a 2.8M English digital OCR model. For more details, please refer to the PP-OCR technical article (https://arxiv.org/abs/2009.09941).
 
-## PP-OCR Pipeline
+[2] On the basis of PP-OCR, PP-OCRv2 is further optimized in five aspects. The detection model adopts CML(Collaborative Mutual Learning) knowledge distillation strategy and CopyPaste data expansion strategy. The recognition model adopts LCNet lightweight backbone network, U-DML knowledge distillation strategy and enhanced CTC loss function improvement (as shown in the red box above), which further improves the inference speed and prediction effect. For more details, please refer to the technical report of PP-OCRv2 (arXiv link is coming soon).
 
-<div align="center">
-    <img src="./doc/ppocr_framework.png" width="800">
-</div>
 
-PP-OCR is a practical ultra-lightweight OCR system. It is mainly composed of three parts: DB text detection[2], detection frame correction and CRNN text recognition[7]. The system adopts 19 effective strategies from 8 aspects including backbone network selection and adjustment, prediction head design, data augmentation, learning rate transformation strategy, regularization parameter selection, pre-training model use, and automatic model tailoring and quantization to optimize and slim down the models of each module. The final results are an ultra-lightweight Chinese and English OCR model with an overall size of 3.5M and a 2.8M English digital OCR model. For more details, please refer to the PP-OCR technical article (https://arxiv.org/abs/2009.09941). Besides, The implementation of the FPGM Pruner [8] and PACT quantization [9] is based on [PaddleSlim](https://github.com/PaddlePaddle/PaddleSlim).
 
 
 ## Visualization [more](./doc/doc_en/visualization_en.md)
@@ -149,7 +174,7 @@ PP-OCR is a practical ultra-lightweight OCR system. It is mainly composed of thr
 
 
 <a name="language_requests"></a>
-## Guideline for new language requests
+## Guideline for New Language Requests
 
 If you want to request a new language support, a PR with 2 following files are needed：
 

README_ch.md

 [English](README.md) | 简体中文
 
+<p align="center">
+ <img src="./doc/PaddleOCR_log.png" align="middle" width = "600"/>
+<p align="center">
+
+
+------------------------------------------------------------------------------------------
+
+<p align="left">
+    <a href="./LICENSE"><img src="https://img.shields.io/badge/license-Apache%202-dfd.svg"></a>
+    <a href="https://github.com/PaddlePaddle/PaddleOCR/releases"><img src="https://img.shields.io/github/v/release/PaddlePaddle/PaddleOCR?color=ffa"></a>
+    <a href=""><img src="https://img.shields.io/badge/python-3.7+-aff.svg"></a>
+    <a href=""><img src="https://img.shields.io/badge/os-linux%2C%20win%2C%20mac-pink.svg"></a>
+    <a href=""><img src="https://img.shields.io/pypi/format/PaddleOCR?color=c77"></a>
+    <a href="https://github.com/PaddlePaddle/PaddleOCR/graphs/contributors"><img src="https://img.shields.io/github/contributors/PaddlePaddle/PaddleOCR?color=9ea"></a>
+    <a href="https://pypi.org/project/PaddleOCR/"><img src="https://img.shields.io/pypi/dm/PaddleOCR?color=9cf"></a>
+    <a href="https://github.com/PaddlePaddle/PaddleOCR/stargazers"><img src="https://img.shields.io/github/stars/PaddlePaddle/PaddleOCR?color=ccf"></a>
+</p>
+
 ## 简介
+
 PaddleOCR旨在打造一套丰富、领先、且实用的OCR工具库，助力使用者训练出更好的模型，并应用落地。
-## 注意
-PaddleOCR同时支持动态图与静态图两种编程范式
-- 动态图版本：dygraph分支（默认），需将paddle版本升级至2.0.0（[快速安装](./doc/doc_ch/installation.md)）
-- 静态图版本：develop分支
 
 **近期更新**
-- 【预告】 PaddleOCR研发团队对最新发版内容技术深入解读，4月13日晚上19:00，[直播地址](https://live.bilibili.com/21689802)
-- 2021.4.8 release 2.1版本，新增AAAI 2021论文[端到端识别算法PGNet](./doc/doc_ch/pgnet.md)开源，[多语言模型](./doc/doc_ch/multi_languages.md)支持种类增加到80+。
-- 2021.2.1 [FAQ](./doc/doc_ch/FAQ.md)新增5个高频问题，总数162个，每周一都会更新，欢迎大家持续关注。
-- 2021.1.21 更新多语言识别模型，目前支持语种超过27种，包括中文简体、中文繁体、英文、法文、德文、韩文、日文、意大利文、西班牙文、葡萄牙文、俄罗斯文、阿拉伯文等，后续计划可以参考[多语言研发计划](https://github.com/PaddlePaddle/PaddleOCR/issues/1048)
-- 2020.12.15 更新数据合成工具[Style-Text](./StyleText/README_ch.md)，可以批量合成大量与目标场景类似的图像，在多个场景验证，效果明显提升。
-- 2020.11.25 更新半自动标注工具[PPOCRLabel](./PPOCRLabel/README_ch.md)，辅助开发者高效完成标注任务，输出格式与PP-OCR训练任务完美衔接。
-- 2020.9.22 更新PP-OCR技术文章，https://arxiv.org/abs/2009.09941
-- [More](./doc/doc_ch/update.md)
-
 
+- PaddleOCR研发团队对最新发版内容技术深入解读，9月8日晚上20:15，[直播地址](https://live.bilibili.com/21689802)。
+- 2021.9.7 发布PaddleOCR v2.3，发布[PP-OCRv2](#PP-OCRv2)，CPU推理速度相比于PP-OCR server提升220%；效果相比于PP-OCR mobile 提升7%。
+- 2021.8.3 发布PaddleOCR v2.2，新增文档结构分析[PP-Structure](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.2/ppstructure/README_ch.md)工具包，支持版面分析与表格识别（含Excel导出）。
+- 2021.6.29 [FAQ](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.2/doc/doc_ch/FAQ.md)新增5个高频问题，总数248个，每周一都会更新，欢迎大家持续关注。
+- 2021.4.8 release 2.1版本，新增AAAI 2021论文[端到端识别算法PGNet](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.2/doc/doc_ch/pgnet.md)开源，[多语言模型](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.2/doc/doc_ch/multi_languages.md)支持种类增加到80+。
+- [More](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.2/doc/doc_ch/update.md)
 
 ## 特性
 
-- PPOCR系列高质量预训练模型，准确的识别效果
-    - 超轻量ppocr_mobile移动端系列：检测（3.0M）+方向分类器（1.4M）+ 识别（5.0M）= 9.4M
-    - 通用ppocr_server系列：检测（47.1M）+方向分类器（1.4M）+ 识别（94.9M）= 143.4M
+- PP-OCR系列高质量预训练模型，准确的识别效果
+    - 超轻量PP-OCRv2系列：检测（3.1M）+ 方向分类器（1.4M）+ 识别（8.5M）= 13.0M
+    - 超轻量PP-OCR mobile移动端系列：检测（3.0M）+方向分类器（1.4M）+ 识别（5.0M）= 9.4M
+    - 通用PPOCR server系列：检测（47.1M）+方向分类器（1.4M）+ 识别（94.9M）= 143.4M
     - 支持中英文数字组合识别、竖排文本识别、长文本识别
     - 支持多语言识别：韩语、日语、德语、法语
 - 丰富易用的OCR相关工具组件
     - 半自动数据标注工具PPOCRLabel：支持快速高效的数据标注
     - 数据合成工具Style-Text：批量合成大量与目标场景类似的图像
+    - 文档分析能力PP-Structure：版面分析与表格识别
 - 支持用户自定义训练，提供丰富的预测推理部署方案
 - 支持PIP快速安装使用
 - 可运行于Linux、Windows、MacOS等多种系统
@@ -40,14 +54,14 @@ PaddleOCR同时支持动态图与静态图两种编程范式
     <img src="doc/imgs_results/ch_ppocr_mobile_v2.0/00018069.jpg" width="800">
 </div>
 
-上图是通用ppocr_server模型效果展示，更多效果图请见[效果展示页面](./doc/doc_ch/visualization.md)。
+上图是通用PP-OCR server模型效果展示，更多效果图请见[效果展示页面](./doc/doc_ch/visualization.md)。
 
 <a name="欢迎加入PaddleOCR技术交流群"></a>
 ## 欢迎加入PaddleOCR技术交流群
 - 微信扫描二维码加入官方交流群，获得更高效的问题答疑，与各行各业开发者充分交流，期待您的加入。
 
 <div align="center">
-<img src="https://raw.githubusercontent.com/PaddlePaddle/PaddleOCR/release/2.0/doc/joinus.PNG"  width = "200" height = "200" />
+<img src="https://raw.githubusercontent.com/PaddlePaddle/PaddleOCR/dygraph/doc/joinus.PNG"  width = "200" height = "200" />
 </div>
 
 ## 快速体验
@@ -63,71 +77,79 @@ PaddleOCR同时支持动态图与静态图两种编程范式
 - 代码体验：从[快速安装](./doc/doc_ch/quickstart.md) 开始
 
 <a name="模型下载"></a>
-## PP-OCR 2.0系列模型列表（更新中）
-**说明** ：2.0版模型和[1.1版模型](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/doc/doc_ch/models_list.md)的主要区别在于动态图训练vs.静态图训练，模型性能上无明显差距。
+## PP-OCR系列模型列表（更新中）
+
 | 模型简介     | 模型名称     |推荐场景          | 检测模型 | 方向分类器 | 识别模型 |
 | ------------ | --------------- | ----------------|---- | ---------- | -------- |
-| 中英文超轻量OCR模型（9.4M） | ch_ppocr_mobile_v2.0_xx |移动端&服务器端|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_train.tar)|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_train.tar) |[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_pre.tar)      |
-| 中英文通用OCR模型（143.4M）   |ch_ppocr_server_v2.0_xx|服务器端 |[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_infer.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_train.tar)    |[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_train.tar)    |[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_infer.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_pre.tar)  |  
+| 中英文超轻量PP-OCRv2模型（13.0M） |  ch_PP-OCRv2_xx |移动端&服务器端|[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_det_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_det_distill_train.tar)| [推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_train.tar) |[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_train.tar)|
+| 中英文超轻量PP-OCR mobile模型（9.4M） | ch_ppocr_mobile_v2.0_xx |移动端&服务器端|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_train.tar)|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_train.tar) |[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_pre.tar)      |
+| 中英文通用PP-OCR server模型（143.4M）   |ch_ppocr_server_v2.0_xx|服务器端 |[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_infer.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_train.tar)    |[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_train.tar)    |[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_infer.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_pre.tar)  |  
 
-更多模型下载（包括多语言），可以参考[PP-OCR v2.0 系列模型下载](./doc/doc_ch/models_list.md)
+更多模型下载（包括多语言），可以参考[PP-OCR 系列模型下载](./doc/doc_ch/models_list.md)
 
 ## 文档教程
-- [快速安装](./doc/doc_ch/installation.md)
-- [中文OCR模型快速使用](./doc/doc_ch/quickstart.md)
-- [多语言OCR模型快速使用](./doc/doc_ch/multi_languages.md)
-- [代码组织结构](./doc/doc_ch/tree.md)
-- 算法介绍
-    - [文本检测](./doc/doc_ch/algorithm_overview.md)
-    - [文本识别](./doc/doc_ch/algorithm_overview.md)
-    - [PP-OCR Pipline](#PP-OCR)
+- [运行环境准备](./doc/doc_ch/environment.md)
+- [快速开始（中英文/多语言/文档分析）](./doc/doc_ch/quickstart.md)
+- [PaddleOCR全景图与项目克隆](./doc/doc_ch/paddleOCR_overview.md)
+- PP-OCR产业落地：从训练到部署
+    - [PP-OCR模型与配置文件](./doc/doc_ch/models_and_config.md)
+        - [PP-OCR模型下载](./doc/doc_ch/models_list.md)
+        - [配置文件内容与生成](./doc/doc_ch/config.md)
+        - [PP-OCR模型库快速推理](./doc/doc_ch/inference_ppocr.md)
+    - [PP-OCR模型训练](./doc/doc_ch/training.md)
+        - [文本检测](./doc/doc_ch/detection.md)
+        - [文本识别](./doc/doc_ch/recognition.md)
+        - [文本方向分类器](./doc/doc_ch/angle_class.md)
+        - [配置文件内容与生成](./doc/doc_ch/config.md)
+    - PP-OCR模型推理部署
+        - [基于C++预测引擎推理](./deploy/cpp_infer/readme.md)
+        - [服务化部署](./deploy/pdserving/README_CN.md)
+        - [端侧部署](./deploy/lite/readme.md)
+        - [Benchmark](./doc/doc_ch/benchmark.md)
+- [PP-Structure信息提取](./ppstructure/README_ch.md)
+    - [版面分析](./ppstructure/layout/README_ch.md)
+    - [表格识别](./ppstructure/table/README_ch.md)
+- 数据标注与合成
+    - [半自动标注工具PPOCRLabel](./PPOCRLabel/README_ch.md)
+    - [数据合成工具Style-Text](./StyleText/README_ch.md)
+    - [其它数据标注工具](./doc/doc_ch/data_annotation.md)
+    - [其它数据合成工具](./doc/doc_ch/data_synthesis.md)
+- OCR学术圈
+    - [两阶段模型介绍与下载](./doc/doc_ch/algorithm_overview.md)
     - [端到端PGNet算法](./doc/doc_ch/pgnet.md)
-- 模型训练/评估
-    - [文本检测](./doc/doc_ch/detection.md)
-    - [文本识别](./doc/doc_ch/recognition.md)
-    - [方向分类器](./doc/doc_ch/angle_class.md)
-    - [yml参数配置文件介绍](./doc/doc_ch/config.md)
-- 预测部署
-    - [基于pip安装whl包快速推理](./doc/doc_ch/whl.md)
     - [基于Python脚本预测引擎推理](./doc/doc_ch/inference.md)
-    - [基于C++预测引擎推理](./deploy/cpp_infer/readme.md)
-    - [服务化部署](./deploy/pdserving/README_CN.md)
-    - [端侧部署](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/deploy/lite/readme.md)
-    - [Benchmark](./doc/doc_ch/benchmark.md)
 - 数据集
     - [通用中英文OCR数据集](./doc/doc_ch/datasets.md)
     - [手写中文OCR数据集](./doc/doc_ch/handwritten_datasets.md)
     - [垂类多语言OCR数据集](./doc/doc_ch/vertical_and_multilingual_datasets.md)
-- 数据标注与合成
-    - [半自动标注工具PPOCRLabel](./PPOCRLabel/README_ch.md)
-    - [数据合成工具Style-Text](./StyleText/README_ch.md)
-    - [其它数据标注工具](./doc/doc_ch/data_annotation.md)
-    - [其它数据合成工具](./doc/doc_ch/data_synthesis.md)
 - [效果展示](#效果展示)
 - FAQ
     - [【精选】OCR精选10个问题](./doc/doc_ch/FAQ.md)
-    - [【理论篇】OCR通用32个问题](./doc/doc_ch/FAQ.md)
-    - [【实战篇】PaddleOCR实战110个问题](./doc/doc_ch/FAQ.md)
+    - [【理论篇】OCR通用50个问题](./doc/doc_ch/FAQ.md)
+    - [【实战篇】PaddleOCR实战183个问题](./doc/doc_ch/FAQ.md)
 - [技术交流群](#欢迎加入PaddleOCR技术交流群)
 - [参考文献](./doc/doc_ch/reference.md)
 - [许可证书](#许可证书)
 - [贡献代码](#贡献代码)
+- [代码组织结构](./doc/doc_ch/tree.md)
+
 
+<a name="PP-OCRv2"></a>
 
-<a name="PP-OCR"></a>
-## PP-OCR Pipline
+## PP-OCRv2 Pipeline
 <div align="center">
-    <img src="./doc/ppocr_framework.png" width="800">
+    <img src="./doc/ppocrv2_framework.jpg" width="800">
 </div>
 
-PP-OCR是一个实用的超轻量OCR系统。主要由DB文本检测[2]、检测框矫正和CRNN文本识别三部分组成[7]。该系统从骨干网络选择和调整、预测头部的设计、数据增强、学习率变换策略、正则化参数选择、预训练模型使用以及模型自动裁剪量化8个方面，采用19个有效策略，对各个模块的模型进行效果调优和瘦身，最终得到整体大小为3.5M的超轻量中英文OCR和2.8M的英文数字OCR。更多细节请参考PP-OCR技术方案 https://arxiv.org/abs/2009.09941 。其中FPGM裁剪器[8]和PACT量化[9]的实现可以参考[PaddleSlim](https://github.com/PaddlePaddle/PaddleSlim)。
+[1] PP-OCR是一个实用的超轻量OCR系统。主要由DB文本检测、检测框矫正和CRNN文本识别三部分组成。该系统从骨干网络选择和调整、预测头部的设计、数据增强、学习率变换策略、正则化参数选择、预训练模型使用以及模型自动裁剪量化8个方面，采用19个有效策略，对各个模块的模型进行效果调优和瘦身(如绿框所示)，最终得到整体大小为3.5M的超轻量中英文OCR和2.8M的英文数字OCR。更多细节请参考PP-OCR技术方案 https://arxiv.org/abs/2009.09941
+
+[2] PP-OCRv2在PP-OCR的基础上，进一步在5个方面重点优化，检测模型采用CML协同互学习知识蒸馏策略和CopyPaste数据增广策略；识别模型采用LCNet轻量级骨干网络、UDML 改进知识蒸馏策略和Enhanced CTC loss损失函数改进（如上图红框所示），进一步在推理速度和预测效果上取得明显提升。更多细节请参考PP-OCR技术方案（arxiv链接生成中）。
+
 
 <a name="效果展示"></a>
 ## 效果展示 [more](./doc/doc_ch/visualization.md)
 - 中文模型
 <div align="center">
-    <img src="./doc/imgs_results/ch_ppocr_mobile_v2.0/test_add_91.jpg" width="800">
-    <img src="./doc/imgs_results/ch_ppocr_mobile_v2.0/00015504.jpg" width="800">
     <img src="./doc/imgs_results/ch_ppocr_mobile_v2.0/00056221.jpg" width="800">
     <img src="./doc/imgs_results/ch_ppocr_mobile_v2.0/rotate_00052204.jpg" width="800">
 </div>

StyleText/engine/text_drawers.py

@@ -66,6 +66,7 @@ class StdTextDrawer(object):
                     corpus_list.append(corpus[0:i])
                     text_input_list.append(text_input)
                     corpus = corpus[i:]
+                    i = 0
                     break
                 draw.text((char_x, 2), char_i, fill=(0, 0, 0), font=font)
                 char_x += char_size
@@ -78,7 +79,6 @@ class StdTextDrawer(object):
 
                 corpus_list.append(corpus[0:i])
                 text_input_list.append(text_input)
-                corpus = corpus[i:]
                 break
 
         return corpus_list, text_input_list

__init__.py

@@ -11,7 +11,8 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
+import paddleocr
+from .paddleocr import *
 
-__all__ = ['PaddleOCR', 'draw_ocr']
-from .paddleocr import PaddleOCR
-from .tools.infer.utility import draw_ocr
+__version__ = paddleocr.VERSION
+__all__ = ['PaddleOCR', 'PPStructure', 'draw_ocr', 'draw_structure_result', 'save_structure_res','download_with_progressbar']

benchmark/analysis.py

+# copyright (c) 2019 PaddlePaddle Authors. All Rights Reserve.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import print_function
+
+import argparse
+import json
+import os
+import re
+import traceback
+
+
+def parse_args():
+    parser = argparse.ArgumentParser(description=__doc__)
+    parser.add_argument(
+        "--filename", type=str, help="The name of log which need to analysis.")
+    parser.add_argument(
+        "--log_with_profiler", type=str, help="The path of train log with profiler")
+    parser.add_argument(
+        "--profiler_path", type=str, help="The path of profiler timeline log.")
+    parser.add_argument(
+        "--keyword", type=str, help="Keyword to specify analysis data")
+    parser.add_argument(
+        "--separator", type=str, default=None, help="Separator of different field in log")
+    parser.add_argument(
+        '--position', type=int, default=None, help='The position of data field')
+    parser.add_argument(
+        '--range', type=str, default="", help='The range of data field to intercept')
+    parser.add_argument(
+        '--base_batch_size', type=int, help='base_batch size on gpu')
+    parser.add_argument(
+        '--skip_steps', type=int, default=0, help='The number of steps to be skipped')
+    parser.add_argument(
+        '--model_mode', type=int, default=-1, help='Analysis mode, default value is -1')
+    parser.add_argument(
+        '--ips_unit', type=str, default=None, help='IPS unit')
+    parser.add_argument(
+        '--model_name', type=str, default=0, help='training model_name, transformer_base')
+    parser.add_argument(
+        '--mission_name', type=str, default=0, help='training mission name')
+    parser.add_argument(
+        '--direction_id', type=int, default=0, help='training direction_id')
+    parser.add_argument(
+        '--run_mode', type=str, default="sp", help='multi process or single process')
+    parser.add_argument(
+        '--index', type=int, default=1, help='{1: speed, 2:mem, 3:profiler, 6:max_batch_size}')
+    parser.add_argument(
+        '--gpu_num', type=int, default=1, help='nums of training gpus')
+    args = parser.parse_args()
+    args.separator = None if args.separator == "None" else args.separator
+    return args
+
+
+def _is_number(num):
+    pattern = re.compile(r'^[-+]?[-0-9]\d*\.\d*|[-+]?\.?[0-9]\d*$')
+    result = pattern.match(num)
+    if result:
+        return True
+    else:
+        return False
+
+
+class TimeAnalyzer(object):
+    def __init__(self, filename, keyword=None, separator=None, position=None, range="-1"):
+        if filename is None:
+            raise Exception("Please specify the filename!")
+
+        if keyword is None:
+            raise Exception("Please specify the keyword!")
+
+        self.filename = filename
+        self.keyword = keyword
+        self.separator = separator
+        self.position = position
+        self.range = range
+        self.records = None
+        self._distil()
+
+    def _distil(self):
+        self.records = []
+        with open(self.filename, "r") as f_object:
+            lines = f_object.readlines()
+            for line in lines:
+                if self.keyword not in line:
+                    continue
+                try:
+                    result = None
+
+                    # Distil the string from a line.
+                    line = line.strip()
+                    line_words = line.split(self.separator) if self.separator else line.split()
+                    if args.position:
+                        result = line_words[self.position]
+                    else:
+                        # Distil the string following the keyword.
+                        for i in range(len(line_words) - 1):
+                            if line_words[i] == self.keyword:
+                                result = line_words[i + 1]
+                                break
+    
+                    # Distil the result from the picked string.
+                    if not self.range:
+                        result = result[0:]
+                    elif _is_number(self.range):
+                        result = result[0: int(self.range)]
+                    else:
+                        result = result[int(self.range.split(":")[0]): int(self.range.split(":")[1])]
+                    self.records.append(float(result))
+                except Exception as exc:
+                    print("line is: {}; separator={}; position={}".format(line, self.separator, self.position))
+
+        print("Extract {} records: separator={}; position={}".format(len(self.records), self.separator, self.position))
+
+    def _get_fps(self, mode, batch_size, gpu_num, avg_of_records, run_mode, unit=None):
+        if mode == -1 and run_mode == 'sp':
+            assert unit, "Please set the unit when mode is -1."
+            fps = gpu_num * avg_of_records
+        elif mode == -1 and run_mode == 'mp':
+            assert unit, "Please set the unit when mode is -1."
+            fps = gpu_num * avg_of_records #temporarily, not used now
+            print("------------this is mp")
+        elif mode == 0:
+            # s/step -> samples/s
+            fps = (batch_size * gpu_num) / avg_of_records
+            unit = "samples/s"
+        elif mode == 1:
+            # steps/s -> steps/s
+            fps = avg_of_records
+            unit = "steps/s"
+        elif mode == 2:
+            # s/step -> steps/s
+            fps = 1 / avg_of_records
+            unit = "steps/s"
+        elif mode == 3:
+            # steps/s -> samples/s
+            fps = batch_size * gpu_num * avg_of_records
+            unit = "samples/s"
+        elif mode == 4:
+            # s/epoch -> s/epoch
+            fps = avg_of_records
+            unit = "s/epoch"
+        else:
+            ValueError("Unsupported analysis mode.")
+
+        return fps, unit
+
+    def analysis(self, batch_size, gpu_num=1, skip_steps=0, mode=-1, run_mode='sp', unit=None):
+        if batch_size <= 0:
+            print("base_batch_size should larger than 0.")
+            return 0, ''
+
+        if len(self.records) <= skip_steps:  # to address the condition which item of log equals to skip_steps
+            print("no records")
+            return 0, ''
+
+        sum_of_records = 0
+        sum_of_records_skipped = 0
+        skip_min = self.records[skip_steps]
+        skip_max = self.records[skip_steps]
+
+        count = len(self.records)
+        for i in range(count):
+            sum_of_records += self.records[i]
+            if i >= skip_steps:
+                sum_of_records_skipped += self.records[i]
+                if self.records[i] < skip_min:
+                    skip_min = self.records[i]
+                if self.records[i] > skip_max:
+                    skip_max = self.records[i]
+
+        avg_of_records = sum_of_records / float(count)
+        avg_of_records_skipped = sum_of_records_skipped / float(count - skip_steps)
+
+        fps, fps_unit = self._get_fps(mode, batch_size, gpu_num, avg_of_records, run_mode, unit)
+        fps_skipped, _ = self._get_fps(mode, batch_size, gpu_num, avg_of_records_skipped, run_mode, unit)
+        if mode == -1:
+            print("average ips of %d steps, skip 0 step:" % count)
+            print("\tAvg: %.3f %s" % (avg_of_records, fps_unit))
+            print("\tFPS: %.3f %s" % (fps, fps_unit))
+            if skip_steps > 0:
+                print("average ips of %d steps, skip %d steps:" % (count, skip_steps))
+                print("\tAvg: %.3f %s" % (avg_of_records_skipped, fps_unit))
+                print("\tMin: %.3f %s" % (skip_min, fps_unit))
+                print("\tMax: %.3f %s" % (skip_max, fps_unit))
+                print("\tFPS: %.3f %s" % (fps_skipped, fps_unit))
+        elif mode == 1 or mode == 3:
+            print("average latency of %d steps, skip 0 step:" % count)
+            print("\tAvg: %.3f steps/s" % avg_of_records)
+            print("\tFPS: %.3f %s" % (fps, fps_unit))
+            if skip_steps > 0:
+                print("average latency of %d steps, skip %d steps:" % (count, skip_steps))
+                print("\tAvg: %.3f steps/s" % avg_of_records_skipped)
+                print("\tMin: %.3f steps/s" % skip_min)
+                print("\tMax: %.3f steps/s" % skip_max)
+                print("\tFPS: %.3f %s" % (fps_skipped, fps_unit))
+        elif mode == 0 or mode == 2:
+            print("average latency of %d steps, skip 0 step:" % count)
+            print("\tAvg: %.3f s/step" % avg_of_records)
+            print("\tFPS: %.3f %s" % (fps, fps_unit))
+            if skip_steps > 0:
+                print("average latency of %d steps, skip %d steps:" % (count, skip_steps))
+                print("\tAvg: %.3f s/step" % avg_of_records_skipped)
+                print("\tMin: %.3f s/step" % skip_min)
+                print("\tMax: %.3f s/step" % skip_max)
+                print("\tFPS: %.3f %s" % (fps_skipped, fps_unit))
+
+        return round(fps_skipped, 3), fps_unit
+
+
+if __name__ == "__main__":
+    args = parse_args()
+    run_info = dict()
+    run_info["log_file"] = args.filename
+    run_info["model_name"] = args.model_name
+    run_info["mission_name"] = args.mission_name
+    run_info["direction_id"] = args.direction_id
+    run_info["run_mode"] = args.run_mode
+    run_info["index"] = args.index
+    run_info["gpu_num"] = args.gpu_num
+    run_info["FINAL_RESULT"] = 0
+    run_info["JOB_FAIL_FLAG"] = 0
+
+    try:
+        if args.index == 1:
+            if args.gpu_num == 1:
+                run_info["log_with_profiler"] = args.log_with_profiler
+                run_info["profiler_path"] = args.profiler_path
+            analyzer = TimeAnalyzer(args.filename, args.keyword, args.separator, args.position, args.range)
+            run_info["FINAL_RESULT"], run_info["UNIT"] = analyzer.analysis(
+                batch_size=args.base_batch_size,
+                gpu_num=args.gpu_num,
+                skip_steps=args.skip_steps,
+                mode=args.model_mode,
+                run_mode=args.run_mode,
+                unit=args.ips_unit)
+            try:
+                if int(os.getenv('job_fail_flag')) == 1 or int(run_info["FINAL_RESULT"]) == 0:
+                    run_info["JOB_FAIL_FLAG"] = 1
+            except:
+                pass
+        elif args.index == 3:
+            run_info["FINAL_RESULT"] = {}
+            records_fo_total = TimeAnalyzer(args.filename, 'Framework overhead', None, 3, '').records
+            records_fo_ratio = TimeAnalyzer(args.filename, 'Framework overhead', None, 5).records
+            records_ct_total = TimeAnalyzer(args.filename, 'Computation time', None, 3, '').records
+            records_gm_total = TimeAnalyzer(args.filename, 'GpuMemcpy                Calls', None, 4, '').records
+            records_gm_ratio = TimeAnalyzer(args.filename, 'GpuMemcpy                Calls', None, 6).records
+            records_gmas_total = TimeAnalyzer(args.filename, 'GpuMemcpyAsync         Calls', None, 4, '').records
+            records_gms_total = TimeAnalyzer(args.filename, 'GpuMemcpySync          Calls', None, 4, '').records
+            run_info["FINAL_RESULT"]["Framework_Total"] = records_fo_total[0] if records_fo_total else 0
+            run_info["FINAL_RESULT"]["Framework_Ratio"] = records_fo_ratio[0] if records_fo_ratio else 0
+            run_info["FINAL_RESULT"]["ComputationTime_Total"] = records_ct_total[0] if records_ct_total else 0
+            run_info["FINAL_RESULT"]["GpuMemcpy_Total"] = records_gm_total[0] if records_gm_total else 0
+            run_info["FINAL_RESULT"]["GpuMemcpy_Ratio"] = records_gm_ratio[0] if records_gm_ratio else 0
+            run_info["FINAL_RESULT"]["GpuMemcpyAsync_Total"] = records_gmas_total[0] if records_gmas_total else 0
+            run_info["FINAL_RESULT"]["GpuMemcpySync_Total"] = records_gms_total[0] if records_gms_total else 0
+        else:
+            print("Not support!")
+    except Exception:
+            traceback.print_exc()
+    print("{}".format(json.dumps(run_info)))  # it's required, for the log file path  insert to the database
+

benchmark/readme.md

+
+# PaddleOCR DB/EAST 算法训练benchmark测试
+
+PaddleOCR/benchmark目录下的文件用于获取并分析训练日志。
+训练采用icdar2015数据集，包括1000张训练图像和500张测试图像。模型配置采用resnet18_vd作为backbone，分别训练batch_size=8和batch_size=16的情况。
+
+## 运行训练benchmark
+
+benchmark/run_det.sh 中包含了三个过程：
+- 安装依赖
+- 下载数据
+- 执行训练
+- 日志分析获取IPS
+
+在执行训练部分，会执行单机单卡（默认0号卡）单机多卡训练，并分别执行batch_size=8和batch_size=16的情况。所以执行完后，每种模型会得到4个日志文件。
+
+run_det.sh 执行方式如下:
+
+```
+# cd PaddleOCR/
+bash benchmark/run_det.sh 
+```
+
+以DB为例，将得到四个日志文件，如下：
+```
+det_res18_db_v2.0_sp_bs16_fp32_1
+det_res18_db_v2.0_sp_bs8_fp32_1
+det_res18_db_v2.0_mp_bs16_fp32_1
+det_res18_db_v2.0_mp_bs8_fp32_1
+```
+
+
+
+

benchmark/run_benchmark_det.sh

+#!/usr/bin/env bash
+set -xe
+# 运行示例：CUDA_VISIBLE_DEVICES=0 bash run_benchmark.sh ${run_mode} ${bs_item} ${fp_item} 500 ${model_mode}
+# 参数说明
+function _set_params(){
+    run_mode=${1:-"sp"}          # 单卡sp|多卡mp
+    batch_size=${2:-"64"}
+    fp_item=${3:-"fp32"}        # fp32|fp16
+    max_iter=${4:-"500"}       # 可选，如果需要修改代码提前中断
+    model_name=${5:-"model_name"}
+    run_log_path=${TRAIN_LOG_DIR:-$(pwd)}  # TRAIN_LOG_DIR 后续QA设置该参数
+
+#   以下不用修改   
+    device=${CUDA_VISIBLE_DEVICES//,/ }
+    arr=(${device})
+    num_gpu_devices=${#arr[*]}
+    log_file=${run_log_path}/${model_name}_${run_mode}_bs${batch_size}_${fp_item}_${num_gpu_devices}
+}
+function _train(){
+    echo "Train on ${num_gpu_devices} GPUs"
+    echo "current CUDA_VISIBLE_DEVICES=$CUDA_VISIBLE_DEVICES, gpus=$num_gpu_devices, batch_size=$batch_size"
+
+    train_cmd="-c configs/det/${model_name}.yml -o Train.loader.batch_size_per_card=${batch_size} Global.epoch_num=${max_iter} "   
+    case ${run_mode} in
+      sp) 
+        train_cmd="python3.7 tools/train.py "${train_cmd}""
+        ;;
+      mp)
+        train_cmd="python3.7 -m paddle.distributed.launch --log_dir=./mylog --gpus=$CUDA_VISIBLE_DEVICES tools/train.py ${train_cmd}"
+        ;;
+      *) echo "choose run_mode(sp or mp)"; exit 1;
+    esac
+# 以下不用修改
+    timeout 15m ${train_cmd} > ${log_file} 2>&1
+    if [ $? -ne 0 ];then
+            echo -e "${model_name}, FAIL"
+        export job_fail_flag=1
+    else
+        echo -e "${model_name}, SUCCESS"
+        export job_fail_flag=0
+    fi
+    kill -9 `ps -ef|grep 'python3.7'|awk '{print $2}'`
+
+    if [ $run_mode = "mp" -a -d mylog ]; then
+        rm ${log_file}
+        cp mylog/workerlog.0 ${log_file}
+    fi
+
+    # run log analysis
+    analysis_cmd="python3.7 benchmark/analysis.py --filename ${log_file}  --mission_name ${model_name} --run_mode ${mode} --direction_id 0 --keyword 'ips:' --base_batch_size ${batch_szie} --skip_steps 1 --gpu_num ${num_gpu_devices}  --index 1  --model_mode=-1  --ips_unit=samples/sec"
+    eval $analysis_cmd
+}
+
+_set_params $@
+_train
+

benchmark/run_det.sh

+# 提供可稳定复现性能的脚本，默认在标准docker环境内py37执行： paddlepaddle/paddle:latest-gpu-cuda10.1-cudnn7  paddle=2.1.2  py=37
+# 执行目录: ./PaddleOCR
+# 1 安装该模型需要的依赖 (如需开启优化策略请注明)
+python3.7 -m pip install -r requirements.txt
+# 2 拷贝该模型需要数据、预训练模型
+wget -c  -p ./tain_data/  https://paddleocr.bj.bcebos.com/dygraph_v2.0/test/icdar2015.tar && cd train_data  && tar xf icdar2015.tar && cd ../
+wget -c -p ./pretrain_models/ https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/ResNet50_vd_pretrained.pdparams
+# 3 批量运行（如不方便批量，1，2需放到单个模型中）
+
+model_mode_list=(det_res18_db_v2.0 det_r50_vd_east)
+fp_item_list=(fp32)
+bs_list=(8 16)
+for model_mode in ${model_mode_list[@]}; do
+      for fp_item in ${fp_item_list[@]}; do
+          for bs_item in ${bs_list[@]}; do
+            echo "index is speed, 1gpus, begin, ${model_name}"
+            run_mode=sp
+            CUDA_VISIBLE_DEVICES=0 bash benchmark/run_benchmark_det.sh ${run_mode} ${bs_item} ${fp_item} 10 ${model_mode}     #  (5min)
+            sleep 60
+            echo "index is speed, 8gpus, run_mode is multi_process, begin, ${model_name}"
+            run_mode=mp
+            CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 bash benchmark/run_benchmark_det.sh ${run_mode} ${bs_item} ${fp_item} 10 ${model_mode} 
+            sleep 60
+            done
+      done
+done
+
+

configs/det/ch_PP-OCRv2/ch_PP-OCR_det_cml.yml

+Global:
+  use_gpu: true
+  epoch_num: 1200
+  log_smooth_window: 20
+  print_batch_step: 2
+  save_model_dir: ./output/ch_db_mv3/
+  save_epoch_step: 1200
+  # evaluation is run every 5000 iterations after the 4000th iteration
+  eval_batch_step: [3000, 2000]
+  cal_metric_during_train: False
+  pretrained_model: ./pretrain_models/ch_PP-OCRv2_det_distill_train/best_accuracy
+  checkpoints:
+  save_inference_dir:
+  use_visualdl: False
+  infer_img: doc/imgs_en/img_10.jpg
+  save_res_path: ./output/det_db/predicts_db.txt
+
+Architecture:
+  name: DistillationModel
+  algorithm: Distillation
+  Models:
+    Teacher:
+      freeze_params: true
+      return_all_feats: false
+      model_type: det
+      algorithm: DB
+      Transform:
+      Backbone:
+        name: ResNet
+        layers: 18
+      Neck:
+        name: DBFPN
+        out_channels: 256
+      Head:
+        name: DBHead
+        k: 50
+    Student:
+      freeze_params: false
+      return_all_feats: false
+      model_type: det
+      algorithm: DB
+      Backbone:
+        name: MobileNetV3
+        scale: 0.5
+        model_name: large
+        disable_se: True
+      Neck:
+        name: DBFPN
+        out_channels: 96
+      Head:
+        name: DBHead
+        k: 50
+    Student2:
+      freeze_params: false
+      return_all_feats: false
+      model_type: det
+      algorithm: DB
+      Transform:
+      Backbone:
+        name: MobileNetV3
+        scale: 0.5
+        model_name: large
+        disable_se: True
+      Neck:
+        name: DBFPN
+        out_channels: 96
+      Head:
+        name: DBHead
+        k: 50
+    
+Loss:
+  name: CombinedLoss
+  loss_config_list:
+  - DistillationDilaDBLoss:
+      weight: 1.0
+      model_name_pairs:
+      - ["Student", "Teacher"]
+      - ["Student2", "Teacher"]
+      key: maps
+      balance_loss: true
+      main_loss_type: DiceLoss
+      alpha: 5
+      beta: 10
+      ohem_ratio: 3
+  - DistillationDMLLoss:
+      model_name_pairs:
+      - ["Student", "Student2"]
+      maps_name: "thrink_maps"
+      weight: 1.0
+      # act: None
+      model_name_pairs: ["Student", "Student2"]
+      key: maps
+  - DistillationDBLoss:
+      weight: 1.0
+      model_name_list: ["Student", "Student2"]
+      # key: maps
+      # name: DBLoss
+      balance_loss: true
+      main_loss_type: DiceLoss
+      alpha: 5
+      beta: 10
+      ohem_ratio: 3
+
+
+Optimizer:
+  name: Adam
+  beta1: 0.9
+  beta2: 0.999
+  lr:
+    name: Cosine
+    learning_rate: 0.001
+    warmup_epoch: 2
+  regularizer:
+    name: 'L2'
+    factor: 0
+
+PostProcess:
+  name: DistillationDBPostProcess
+  model_name: ["Student", "Student2", "Teacher"]
+  # key: maps
+  thresh: 0.3
+  box_thresh: 0.6
+  max_candidates: 1000
+  unclip_ratio: 1.5
+
+Metric:
+  name: DistillationMetric
+  base_metric_name: DetMetric
+  main_indicator: hmean
+  key: "Student"
+
+Train:
+  dataset:
+    name: SimpleDataSet
+    data_dir: ./train_data/icdar2015/text_localization/
+    label_file_list:
+      - ./train_data/icdar2015/text_localization/train_icdar2015_label.txt
+    ratio_list: [1.0]
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - DetLabelEncode: # Class handling label
+      - IaaAugment:
+          augmenter_args:
+            - { 'type': Fliplr, 'args': { 'p': 0.5 } }
+            - { 'type': Affine, 'args': { 'rotate': [-10, 10] } }
+            - { 'type': Resize, 'args': { 'size': [0.5, 3] } }
+      - EastRandomCropData:
+          size: [960, 960]
+          max_tries: 50
+          keep_ratio: true
+      - MakeBorderMap:
+          shrink_ratio: 0.4
+          thresh_min: 0.3
+          thresh_max: 0.7
+      - MakeShrinkMap:
+          shrink_ratio: 0.4
+          min_text_size: 8
+      - NormalizeImage:
+          scale: 1./255.
+          mean: [0.485, 0.456, 0.406]
+          std: [0.229, 0.224, 0.225]
+          order: 'hwc'
+      - ToCHWImage:
+      - KeepKeys:
+          keep_keys: ['image', 'threshold_map', 'threshold_mask', 'shrink_map', 'shrink_mask'] # the order of the dataloader list
+  loader:
+    shuffle: True
+    drop_last: False
+    batch_size_per_card: 8
+    num_workers: 4
+
+Eval:
+  dataset:
+    name: SimpleDataSet
+    data_dir: ./train_data/icdar2015/text_localization/
+    label_file_list:
+      - ./train_data/icdar2015/text_localization/test_icdar2015_label.txt
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - DetLabelEncode: # Class handling label
+      - DetResizeForTest:
+#           image_shape: [736, 1280]
+      - NormalizeImage:
+          scale: 1./255.
+          mean: [0.485, 0.456, 0.406]
+          std: [0.229, 0.224, 0.225]
+          order: 'hwc'
+      - ToCHWImage:
+      - KeepKeys:
+          keep_keys: ['image', 'shape', 'polys', 'ignore_tags']
+  loader:
+    shuffle: False
+    drop_last: False
+    batch_size_per_card: 1 # must be 1
+    num_workers: 2

configs/det/ch_PP-OCRv2/ch_PP-OCR_det_distill.yml

+Global:
+  use_gpu: true
+  epoch_num: 1200
+  log_smooth_window: 20
+  print_batch_step: 2
+  save_model_dir: ./output/ch_db_mv3/
+  save_epoch_step: 1200
+  # evaluation is run every 5000 iterations after the 4000th iteration
+  eval_batch_step: [3000, 2000]
+  cal_metric_during_train: False
+  pretrained_model: ./pretrain_models/MobileNetV3_large_x0_5_pretrained
+  checkpoints:
+  save_inference_dir:
+  use_visualdl: False
+  infer_img: doc/imgs_en/img_10.jpg
+  save_res_path: ./output/det_db/predicts_db.txt
+
+Architecture:
+  name: DistillationModel
+  algorithm: Distillation
+  Models:
+    Student:
+      pretrained: ./pretrain_models/MobileNetV3_large_x0_5_pretrained
+      freeze_params: false
+      return_all_feats: false
+      model_type: det
+      algorithm: DB
+      Backbone:
+        name: MobileNetV3
+        scale: 0.5
+        model_name: large
+        disable_se: True
+      Neck:
+        name: DBFPN
+        out_channels: 96
+      Head:
+        name: DBHead
+        k: 50
+    Teacher:
+      pretrained: ./pretrain_models/ch_ppocr_server_v2.0_det_train/best_accuracy
+      freeze_params: true
+      return_all_feats: false
+      model_type: det
+      algorithm: DB
+      Transform:
+      Backbone:
+        name: ResNet
+        layers: 18
+      Neck:
+        name: DBFPN
+        out_channels: 256
+      Head:
+        name: DBHead
+        k: 50
+
+Loss:
+  name: CombinedLoss
+  loss_config_list:
+  - DistillationDilaDBLoss:
+      weight: 1.0
+      model_name_pairs:
+      - ["Student", "Teacher"]
+      key: maps
+      balance_loss: true
+      main_loss_type: DiceLoss
+      alpha: 5
+      beta: 10
+      ohem_ratio: 3
+  - DistillationDBLoss:
+      weight: 1.0
+      model_name_list: ["Student", "Teacher"]
+      # key: maps
+      name: DBLoss
+      balance_loss: true
+      main_loss_type: DiceLoss
+      alpha: 5
+      beta: 10
+      ohem_ratio: 3
+
+Optimizer:
+  name: Adam
+  beta1: 0.9
+  beta2: 0.999
+  lr:
+    name: Cosine
+    learning_rate: 0.001
+    warmup_epoch: 2
+  regularizer:
+    name: 'L2'
+    factor: 0
+
+PostProcess:
+  name: DistillationDBPostProcess
+  model_name: ["Student", "Student2"]
+  key: head_out
+  thresh: 0.3
+  box_thresh: 0.6
+  max_candidates: 1000
+  unclip_ratio: 1.5
+
+Metric:
+  name: DistillationMetric
+  base_metric_name: DetMetric
+  main_indicator: hmean
+  key: "Student"
+
+Train:
+  dataset:
+    name: SimpleDataSet
+    data_dir: ./train_data/icdar2015/text_localization/
+    label_file_list:
+      - ./train_data/icdar2015/text_localization/train_icdar2015_label.txt
+    ratio_list: [1.0]
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - DetLabelEncode: # Class handling label
+      - IaaAugment:
+          augmenter_args:
+            - { 'type': Fliplr, 'args': { 'p': 0.5 } }
+            - { 'type': Affine, 'args': { 'rotate': [-10, 10] } }
+            - { 'type': Resize, 'args': { 'size': [0.5, 3] } }
+      - EastRandomCropData:
+          size: [960, 960]
+          max_tries: 50
+          keep_ratio: true
+      - MakeBorderMap:
+          shrink_ratio: 0.4
+          thresh_min: 0.3
+          thresh_max: 0.7
+      - MakeShrinkMap:
+          shrink_ratio: 0.4
+          min_text_size: 8
+      - NormalizeImage:
+          scale: 1./255.
+          mean: [0.485, 0.456, 0.406]
+          std: [0.229, 0.224, 0.225]
+          order: 'hwc'
+      - ToCHWImage:
+      - KeepKeys:
+          keep_keys: ['image', 'threshold_map', 'threshold_mask', 'shrink_map', 'shrink_mask'] # the order of the dataloader list
+  loader:
+    shuffle: True
+    drop_last: False
+    batch_size_per_card: 8
+    num_workers: 4
+
+Eval:
+  dataset:
+    name: SimpleDataSet
+    data_dir: ./train_data/icdar2015/text_localization/
+    label_file_list:
+      - ./train_data/icdar2015/text_localization/test_icdar2015_label.txt
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - DetLabelEncode: # Class handling label
+      - DetResizeForTest:
+#           image_shape: [736, 1280]
+      - NormalizeImage:
+          scale: 1./255.
+          mean: [0.485, 0.456, 0.406]
+          std: [0.229, 0.224, 0.225]
+          order: 'hwc'
+      - ToCHWImage:
+      - KeepKeys:
+          keep_keys: ['image', 'shape', 'polys', 'ignore_tags']
+  loader:
+    shuffle: False
+    drop_last: False
+    batch_size_per_card: 1 # must be 1
+    num_workers: 2

configs/det/ch_PP-OCRv2/ch_PP-OCR_det_dml.yml

+Global:
+  use_gpu: true
+  epoch_num: 1200
+  log_smooth_window: 20
+  print_batch_step: 2
+  save_model_dir: ./output/ch_db_mv3/
+  save_epoch_step: 1200
+  # evaluation is run every 5000 iterations after the 4000th iteration
+  eval_batch_step: [3000, 2000]
+  cal_metric_during_train: False
+  pretrained_model: ./pretrain_models/MobileNetV3_large_x0_5_pretrained
+  checkpoints:
+  save_inference_dir:
+  use_visualdl: False
+  infer_img: doc/imgs_en/img_10.jpg
+  save_res_path: ./output/det_db/predicts_db.txt
+
+Architecture:
+  name: DistillationModel
+  algorithm: Distillation
+  Models:
+    Student:
+      pretrained: ./pretrain_models/MobileNetV3_large_x0_5_pretrained
+      freeze_params: false
+      return_all_feats: false
+      model_type: det
+      algorithm: DB
+      Backbone:
+        name: MobileNetV3
+        scale: 0.5
+        model_name: large
+        disable_se: True
+      Neck:
+        name: DBFPN
+        out_channels: 96
+      Head:
+        name: DBHead
+        k: 50
+    Student2:
+      pretrained: ./pretrain_models/MobileNetV3_large_x0_5_pretrained
+      freeze_params: false
+      return_all_feats: false
+      model_type: det
+      algorithm: DB
+      Transform:
+      Backbone:
+        name: MobileNetV3
+        scale: 0.5
+        model_name: large
+        disable_se: True
+      Neck:
+        name: DBFPN
+        out_channels: 96
+      Head:
+        name: DBHead
+        k: 50
+
+
+Loss:
+  name: CombinedLoss
+  loss_config_list:
+  - DistillationDMLLoss:
+      model_name_pairs:
+      - ["Student", "Student2"]
+      maps_name: "thrink_maps"
+      weight: 1.0
+      act: "softmax"
+      model_name_pairs: ["Student", "Student2"]
+      key: maps
+  - DistillationDBLoss:
+      weight: 1.0
+      model_name_list: ["Student", "Student2"]
+      # key: maps
+      name: DBLoss
+      balance_loss: true
+      main_loss_type: DiceLoss
+      alpha: 5
+      beta: 10
+      ohem_ratio: 3
+
+
+Optimizer:
+  name: Adam
+  beta1: 0.9
+  beta2: 0.999
+  lr:
+    name: Cosine
+    learning_rate: 0.001
+    warmup_epoch: 2
+  regularizer:
+    name: 'L2'
+    factor: 0
+
+PostProcess:
+  name: DistillationDBPostProcess
+  model_name: ["Student", "Student2"]
+  key: head_out
+  thresh: 0.3
+  box_thresh: 0.6
+  max_candidates: 1000
+  unclip_ratio: 1.5
+
+Metric:
+  name: DistillationMetric
+  base_metric_name: DetMetric
+  main_indicator: hmean
+  key: "Student"
+
+Train:
+  dataset:
+    name: SimpleDataSet
+    data_dir: ./train_data/icdar2015/text_localization/
+    label_file_list:
+      - ./train_data/icdar2015/text_localization/train_icdar2015_label.txt
+    ratio_list: [1.0]
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - DetLabelEncode: # Class handling label
+      - IaaAugment:
+          augmenter_args:
+            - { 'type': Fliplr, 'args': { 'p': 0.5 } }
+            - { 'type': Affine, 'args': { 'rotate': [-10, 10] } }
+            - { 'type': Resize, 'args': { 'size': [0.5, 3] } }
+      - EastRandomCropData:
+          size: [960, 960]
+          max_tries: 50
+          keep_ratio: true
+      - MakeBorderMap:
+          shrink_ratio: 0.4
+          thresh_min: 0.3
+          thresh_max: 0.7
+      - MakeShrinkMap:
+          shrink_ratio: 0.4
+          min_text_size: 8
+      - NormalizeImage:
+          scale: 1./255.
+          mean: [0.485, 0.456, 0.406]
+          std: [0.229, 0.224, 0.225]
+          order: 'hwc'
+      - ToCHWImage:
+      - KeepKeys:
+          keep_keys: ['image', 'threshold_map', 'threshold_mask', 'shrink_map', 'shrink_mask'] # the order of the dataloader list
+  loader:
+    shuffle: True
+    drop_last: False
+    batch_size_per_card: 8
+    num_workers: 4
+
+Eval:
+  dataset:
+    name: SimpleDataSet
+    data_dir: ./train_data/icdar2015/text_localization/
+    label_file_list:
+      - ./train_data/icdar2015/text_localization/test_icdar2015_label.txt
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - DetLabelEncode: # Class handling label
+      - DetResizeForTest:
+#           image_shape: [736, 1280]
+      - NormalizeImage:
+          scale: 1./255.
+          mean: [0.485, 0.456, 0.406]
+          std: [0.229, 0.224, 0.225]
+          order: 'hwc'
+      - ToCHWImage:
+      - KeepKeys:
+          keep_keys: ['image', 'shape', 'polys', 'ignore_tags']
+  loader:
+    shuffle: False
+    drop_last: False
+    batch_size_per_card: 1 # must be 1
+    num_workers: 2

configs/det/ch_PP-OCRv2/ch_PP-OCR_det_student.yml

+Global:
+  use_gpu: true
+  epoch_num: 1200
+  log_smooth_window: 20
+  print_batch_step: 10
+  save_model_dir: ./output/ch_db_mv3/
+  save_epoch_step: 1200
+  # evaluation is run every 5000 iterations after the 4000th iteration
+  eval_batch_step: [0, 400]
+  cal_metric_during_train: False
+  pretrained_model: ./pretrain_models/student.pdparams
+  checkpoints:
+  save_inference_dir:
+  use_visualdl: False
+  infer_img: doc/imgs_en/img_10.jpg
+  save_res_path: ./output/det_db/predicts_db.txt
+
+Architecture:
+  model_type: det
+  algorithm: DB
+  Transform:
+  Backbone:
+    name: MobileNetV3
+    scale: 0.5
+    model_name: large
+    disable_se: True
+  Neck:
+    name: DBFPN
+    out_channels: 96
+  Head:
+    name: DBHead
+    k: 50
+
+Loss:
+  name: DBLoss
+  balance_loss: true
+  main_loss_type: DiceLoss
+  alpha: 5
+  beta: 10
+  ohem_ratio: 3
+
+Optimizer:
+  name: Adam
+  beta1: 0.9
+  beta2: 0.999
+  lr:
+    name: Cosine
+    learning_rate: 0.001
+    warmup_epoch: 2
+  regularizer:
+    name: 'L2'
+    factor: 0
+
+PostProcess:
+  name: DBPostProcess
+  thresh: 0.3
+  box_thresh: 0.6
+  max_candidates: 1000
+  unclip_ratio: 1.5
+
+Metric:
+  name: DetMetric
+  main_indicator: hmean
+
+Train:
+  dataset:
+    name: SimpleDataSet
+    data_dir: ./train_data/icdar2015/text_localization/
+    label_file_list:
+      - ./train_data/icdar2015/text_localization/train_icdar2015_label.txt
+    ratio_list: [1.0]
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - DetLabelEncode: # Class handling label
+      - IaaAugment:
+          augmenter_args:
+            - { 'type': Fliplr, 'args': { 'p': 0.5 } }
+            - { 'type': Affine, 'args': { 'rotate': [-10, 10] } }
+            - { 'type': Resize, 'args': { 'size': [0.5, 3] } }
+      - EastRandomCropData:
+          size: [960, 960]
+          max_tries: 50
+          keep_ratio: true
+      - MakeBorderMap:
+          shrink_ratio: 0.4
+          thresh_min: 0.3
+          thresh_max: 0.7
+      - MakeShrinkMap:
+          shrink_ratio: 0.4
+          min_text_size: 8
+      - NormalizeImage:
+          scale: 1./255.
+          mean: [0.485, 0.456, 0.406]
+          std: [0.229, 0.224, 0.225]
+          order: 'hwc'
+      - ToCHWImage:
+      - KeepKeys:
+          keep_keys: ['image', 'threshold_map', 'threshold_mask', 'shrink_map', 'shrink_mask'] # the order of the dataloader list
+  loader:
+    shuffle: True
+    drop_last: False
+    batch_size_per_card: 8
+    num_workers: 4
+
+Eval:
+  dataset:
+    name: SimpleDataSet
+    data_dir: ./train_data/icdar2015/text_localization/
+    label_file_list:
+      - ./train_data/icdar2015/text_localization/test_icdar2015_label.txt
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - DetLabelEncode: # Class handling label
+      - DetResizeForTest:
+#           image_shape: [736, 1280]
+      - NormalizeImage:
+          scale: 1./255.
+          mean: [0.485, 0.456, 0.406]
+          std: [0.229, 0.224, 0.225]
+          order: 'hwc'
+      - ToCHWImage:
+      - KeepKeys:
+          keep_keys: ['image', 'shape', 'polys', 'ignore_tags']
+  loader:
+    shuffle: False
+    drop_last: False
+    batch_size_per_card: 1 # must be 1
+    num_workers: 2

configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml

@@ -7,11 +7,6 @@ Global:
   save_epoch_step: 1200
   # evaluation is run every 5000 iterations after the 4000th iteration
   eval_batch_step: [3000, 2000]
-  # 1. If pretrained_model is saved in static mode, such as classification pretrained model
-  #    from static branch, load_static_weights must be set as True.
-  # 2. If you want to finetune the pretrained models we provide in the docs,
-  #    you should set load_static_weights as False.
-  load_static_weights: True
   cal_metric_during_train: False
   pretrained_model: ./pretrain_models/MobileNetV3_large_x0_5_pretrained
   checkpoints:

configs/det/ch_ppocr_v2.0/ch_det_res18_db_v2.0.yml

@@ -7,11 +7,6 @@ Global:
   save_epoch_step: 1200
   # evaluation is run every 5000 iterations after the 4000th iteration
   eval_batch_step: [3000, 2000]
-  # 1. If pretrained_model is saved in static mode, such as classification pretrained model
-  #    from static branch, load_static_weights must be set as True.
-  # 2. If you want to finetune the pretrained models we provide in the docs,
-  #    you should set load_static_weights as False.
-  load_static_weights: True
   cal_metric_during_train: False
   pretrained_model: ./pretrain_models/ResNet18_vd_pretrained
   checkpoints:

configs/det/det_mv3_db.yml

@@ -7,11 +7,6 @@ Global:
   save_epoch_step: 1200
   # evaluation is run every 2000 iterations
   eval_batch_step: [0, 2000]
-  # 1. If pretrained_model is saved in static mode, such as classification pretrained model
-  #    from static branch, load_static_weights must be set as True.
-  # 2. If you want to finetune the pretrained models we provide in the docs,
-  #    you should set load_static_weights as False.
-  load_static_weights: True
   cal_metric_during_train: False
   pretrained_model: ./pretrain_models/MobileNetV3_large_x0_5_pretrained
   checkpoints:
@@ -133,4 +128,4 @@ Eval:
     drop_last: False
     batch_size_per_card: 1 # must be 1
     num_workers: 8
-    use_shared_memory: False
\ No newline at end of file
+    use_shared_memory: False

configs/det/det_mv3_east.yml

@@ -7,11 +7,6 @@ Global:
   save_epoch_step: 1000
   # evaluation is run every 5000 iterations after the 4000th iteration
   eval_batch_step: [4000, 5000]
-  # 1. If pretrained_model is saved in static mode, such as classification pretrained model
-  #    from static branch, load_static_weights must be set as True.
-  # 2. If you want to finetune the pretrained models we provide in the docs,
-  #    you should set load_static_weights as False.
-  load_static_weights: True
   cal_metric_during_train: False
   pretrained_model: ./pretrain_models/MobileNetV3_large_x0_5_pretrained
   checkpoints: 

configs/det/det_mv3_pse.yml

+Global:
+  use_gpu: true
+  epoch_num: 600
+  log_smooth_window: 20
+  print_batch_step: 10
+  save_model_dir: ./output/det_mv3_pse/
+  save_epoch_step: 600
+  # evaluation is run every 63 iterations
+  eval_batch_step: [ 0,63 ]
+  cal_metric_during_train: False
+  pretrained_model: ./pretrain_models/MobileNetV3_large_x0_5_pretrained
+  checkpoints: #./output/det_r50_vd_pse_batch8_ColorJitter/best_accuracy
+  save_inference_dir:
+  use_visualdl: False
+  infer_img: doc/imgs_en/img_10.jpg
+  save_res_path: ./output/det_pse/predicts_pse.txt
+
+Architecture:
+  model_type: det
+  algorithm: PSE
+  Transform: null
+  Backbone:
+    name: MobileNetV3
+    scale: 0.5
+    model_name: large
+  Neck:
+    name: FPN
+    out_channels: 96
+  Head:
+    name: PSEHead
+    hidden_dim: 96
+    out_channels: 7
+
+Loss:
+  name: PSELoss
+  alpha: 0.7
+  ohem_ratio: 3
+  kernel_sample_mask: pred
+  reduction: none
+
+Optimizer:
+  name: Adam
+  beta1: 0.9
+  beta2: 0.999
+  lr:
+    name: Step
+    learning_rate: 0.001
+    step_size: 200
+    gamma: 0.1
+  regularizer:
+    name: 'L2'
+    factor: 0.0005
+
+PostProcess:
+  name: PSEPostProcess
+  thresh: 0
+  box_thresh: 0.85
+  min_area: 16
+  box_type: box # 'box' or 'poly'
+  scale: 1
+
+Metric:
+  name: DetMetric
+  main_indicator: hmean
+
+Train:
+  dataset:
+    name: SimpleDataSet
+    data_dir: ./train_data/icdar2015/text_localization/
+    label_file_list:
+      - ./train_data/icdar2015/text_localization/train_icdar2015_label.txt
+    ratio_list: [ 1.0 ]
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - DetLabelEncode: # Class handling label
+      - ColorJitter:
+          brightness: 0.12549019607843137
+          saturation: 0.5
+      - IaaAugment:
+          augmenter_args:
+            - { 'type': Resize, 'args': { 'size': [ 0.5, 3 ] } }
+            - { 'type': Fliplr, 'args': { 'p': 0.5 } }
+            - { 'type': Affine, 'args': { 'rotate': [ -10, 10 ] } }
+      - MakePseGt:
+          kernel_num: 7
+          min_shrink_ratio: 0.4
+          size: 640
+      - RandomCropImgMask:
+          size: [ 640,640 ]
+          main_key: gt_text
+          crop_keys: [ 'image', 'gt_text', 'gt_kernels', 'mask' ]
+      - NormalizeImage:
+          scale: 1./255.
+          mean: [ 0.485, 0.456, 0.406 ]
+          std: [ 0.229, 0.224, 0.225 ]
+          order: 'hwc'
+      - ToCHWImage:
+      - KeepKeys:
+          keep_keys: [ 'image', 'gt_text', 'gt_kernels', 'mask' ] # the order of the dataloader list
+  loader:
+    shuffle: True
+    drop_last: False
+    batch_size_per_card: 16
+    num_workers: 8
+
+Eval:
+  dataset:
+    name: SimpleDataSet
+    data_dir: ./train_data/icdar2015/text_localization/
+    label_file_list:
+      - ./train_data/icdar2015/text_localization/test_icdar2015_label.txt
+    ratio_list: [ 1.0 ]
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - DetLabelEncode: # Class handling label
+      - DetResizeForTest:
+          limit_side_len: 736
+          limit_type: min
+      - NormalizeImage:
+          scale: 1./255.
+          mean: [ 0.485, 0.456, 0.406 ]
+          std: [ 0.229, 0.224, 0.225 ]
+          order: 'hwc'
+      - ToCHWImage:
+      - KeepKeys:
+          keep_keys: [ 'image', 'shape', 'polys', 'ignore_tags' ]
+  loader:
+    shuffle: False
+    drop_last: False
+    batch_size_per_card: 1 # must be 1
+    num_workers: 8
\ No newline at end of file

configs/det/det_r50_vd_db.yml

@@ -7,11 +7,6 @@ Global:
   save_epoch_step: 1200
   # evaluation is run every 2000 iterations
   eval_batch_step: [0,2000]
-  # 1. If pretrained_model is saved in static mode, such as classification pretrained model
-  #    from static branch, load_static_weights must be set as True.
-  # 2. If you want to finetune the pretrained models we provide in the docs,
-  #    you should set load_static_weights as False.
-  load_static_weights: True
   cal_metric_during_train: False
   pretrained_model: ./pretrain_models/ResNet50_vd_ssld_pretrained
   checkpoints:
@@ -103,7 +98,7 @@ Train:
     shuffle: True
     drop_last: False
     batch_size_per_card: 16
-    num_workers: 8
+    num_workers: 4
 
 Eval:
   dataset:
@@ -130,4 +125,4 @@ Eval:
     shuffle: False
     drop_last: False
     batch_size_per_card: 1 # must be 1
-    num_workers: 8
\ No newline at end of file
+    num_workers: 8

configs/det/det_r50_vd_east.yml

@@ -7,13 +7,8 @@ Global:
   save_epoch_step: 1000
   # evaluation is run every 5000 iterations after the 4000th iteration
   eval_batch_step: [4000, 5000]
-  # 1. If pretrained_model is saved in static mode, such as classification pretrained model
-  #    from static branch, load_static_weights must be set as True.
-  # 2. If you want to finetune the pretrained models we provide in the docs,
-  #    you should set load_static_weights as False.
-  load_static_weights: True
   cal_metric_during_train: False
-  pretrained_model: ./pretrain_models/ResNet50_vd_pretrained/
+  pretrained_model: ./pretrain_models/ResNet50_vd_pretrained
   checkpoints: 
   save_inference_dir:
   use_visualdl: False

configs/det/det_r50_vd_pse.yml

+Global:
+  use_gpu: true
+  epoch_num: 600
+  log_smooth_window: 20
+  print_batch_step: 10
+  save_model_dir: ./output/det_r50_vd_pse/
+  save_epoch_step: 600
+  # evaluation is run every 125 iterations
+  eval_batch_step: [ 0,125 ]
+  cal_metric_during_train: False
+  pretrained_model: ./pretrain_models/ResNet50_vd_ssld_pretrained
+  checkpoints: #./output/det_r50_vd_pse_batch8_ColorJitter/best_accuracy
+  save_inference_dir:
+  use_visualdl: False
+  infer_img: doc/imgs_en/img_10.jpg
+  save_res_path: ./output/det_pse/predicts_pse.txt
+
+Architecture:
+  model_type: det
+  algorithm: PSE
+  Transform:
+  Backbone:
+    name: ResNet
+    layers: 50
+  Neck:
+    name: FPN
+    out_channels: 256
+  Head:
+    name: PSEHead
+    hidden_dim: 256
+    out_channels: 7
+
+Loss:
+  name: PSELoss
+  alpha: 0.7
+  ohem_ratio: 3
+  kernel_sample_mask: pred
+  reduction: none
+
+Optimizer:
+  name: Adam
+  beta1: 0.9
+  beta2: 0.999
+  lr:
+    name: Step
+    learning_rate: 0.0001
+    step_size: 200
+    gamma: 0.1
+  regularizer:
+    name: 'L2'
+    factor: 0.0005
+
+PostProcess:
+  name: PSEPostProcess
+  thresh: 0
+  box_thresh: 0.85
+  min_area: 16
+  box_type: box # 'box' or 'poly'
+  scale: 1
+
+Metric:
+  name: DetMetric
+  main_indicator: hmean
+
+Train:
+  dataset:
+    name: SimpleDataSet
+    data_dir: ./train_data/icdar2015/text_localization/
+    label_file_list:
+      - ./train_data/icdar2015/text_localization/train_icdar2015_label.txt
+    ratio_list: [ 1.0 ]
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - DetLabelEncode: # Class handling label
+      - ColorJitter:
+          brightness: 0.12549019607843137
+          saturation: 0.5
+      - IaaAugment:
+          augmenter_args:
+            - { 'type': Resize, 'args': { 'size': [ 0.5, 3 ] } }
+            - { 'type': Fliplr, 'args': { 'p': 0.5 } }
+            - { 'type': Affine, 'args': { 'rotate': [ -10, 10 ] } }
+      - MakePseGt:
+          kernel_num: 7
+          min_shrink_ratio: 0.4
+          size: 640
+      - RandomCropImgMask:
+          size: [ 640,640 ]
+          main_key: gt_text
+          crop_keys: [ 'image', 'gt_text', 'gt_kernels', 'mask' ]
+      - NormalizeImage:
+          scale: 1./255.
+          mean: [ 0.485, 0.456, 0.406 ]
+          std: [ 0.229, 0.224, 0.225 ]
+          order: 'hwc'
+      - ToCHWImage:
+      - KeepKeys:
+          keep_keys: [ 'image', 'gt_text', 'gt_kernels', 'mask' ] # the order of the dataloader list
+  loader:
+    shuffle: True
+    drop_last: False
+    batch_size_per_card: 8
+    num_workers: 8
+
+Eval:
+  dataset:
+    name: SimpleDataSet
+    data_dir: ./train_data/icdar2015/text_localization/
+    label_file_list:
+      - ./train_data/icdar2015/text_localization/test_icdar2015_label.txt
+    ratio_list: [ 1.0 ]
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - DetLabelEncode: # Class handling label
+      - DetResizeForTest:
+          limit_side_len: 736
+          limit_type: min
+      - NormalizeImage:
+          scale: 1./255.
+          mean: [ 0.485, 0.456, 0.406 ]
+          std: [ 0.229, 0.224, 0.225 ]
+          order: 'hwc'
+      - ToCHWImage:
+      - KeepKeys:
+          keep_keys: [ 'image', 'shape', 'polys', 'ignore_tags' ]
+  loader:
+    shuffle: False
+    drop_last: False
+    batch_size_per_card: 1 # must be 1
+    num_workers: 8
\ No newline at end of file

configs/det/det_r50_vd_sast_icdar15.yml

@@ -7,11 +7,6 @@ Global:
   save_epoch_step: 1000
   # evaluation is run every 5000 iterations after the 4000th iteration
   eval_batch_step: [4000, 5000]
-  # 1. If pretrained_model is saved in static mode, such as classification pretrained model
-  #    from static branch, load_static_weights must be set as True.
-  # 2. If you want to finetune the pretrained models we provide in the docs,
-  #    you should set load_static_weights as False.
-  load_static_weights: True
   cal_metric_during_train: False
   pretrained_model: ./pretrain_models/ResNet50_vd_ssld_pretrained/
   checkpoints:

configs/det/det_r50_vd_sast_totaltext.yml

@@ -7,11 +7,6 @@ Global:
   save_epoch_step: 1000
   # evaluation is run every 5000 iterations after the 4000th iteration
   eval_batch_step: [4000, 5000]
-  # 1. If pretrained_model is saved in static mode, such as classification pretrained model
-  #    from static branch, load_static_weights must be set as True.
-  # 2. If you want to finetune the pretrained models we provide in the docs,
-  #    you should set load_static_weights as False.
-  load_static_weights: True
   cal_metric_during_train: False
   pretrained_model: ./pretrain_models/ResNet50_vd_ssld_pretrained/
   checkpoints: 

configs/det/det_res18_db_v2.0.yml

+Global:
+  use_gpu: true
+  epoch_num: 1200
+  log_smooth_window: 20
+  print_batch_step: 2
+  save_model_dir: ./output/ch_db_res18/
+  save_epoch_step: 1200
+  # evaluation is run every 5000 iterations after the 4000th iteration
+  eval_batch_step: [3000, 2000]
+  cal_metric_during_train: False
+  pretrained_model: ./pretrain_models/ResNet18_vd_pretrained
+  checkpoints:
+  save_inference_dir:
+  use_visualdl: False
+  infer_img: doc/imgs_en/img_10.jpg
+  save_res_path: ./output/det_db/predicts_db.txt
+
+Architecture:
+  model_type: det
+  algorithm: DB
+  Transform:
+  Backbone:
+    name: ResNet
+    layers: 18
+    disable_se: True
+  Neck:
+    name: DBFPN
+    out_channels: 256
+  Head:
+    name: DBHead
+    k: 50
+
+Loss:
+  name: DBLoss
+  balance_loss: true
+  main_loss_type: DiceLoss
+  alpha: 5
+  beta: 10
+  ohem_ratio: 3
+
+Optimizer:
+  name: Adam
+  beta1: 0.9
+  beta2: 0.999
+  lr:
+    name: Cosine
+    learning_rate: 0.001
+    warmup_epoch: 2
+  regularizer:
+    name: 'L2'
+    factor: 0
+
+PostProcess:
+  name: DBPostProcess
+  thresh: 0.3
+  box_thresh: 0.6
+  max_candidates: 1000
+  unclip_ratio: 1.5
+
+Metric:
+  name: DetMetric
+  main_indicator: hmean
+
+Train:
+  dataset:
+    name: SimpleDataSet
+    data_dir: ./train_data/icdar2015/text_localization/
+    label_file_list:
+      - ./train_data/icdar2015/text_localization/train_icdar2015_label.txt
+    ratio_list: [1.0]
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - DetLabelEncode: # Class handling label
+      - IaaAugment:
+          augmenter_args:
+            - { 'type': Fliplr, 'args': { 'p': 0.5 } }
+            - { 'type': Affine, 'args': { 'rotate': [-10, 10] } }
+            - { 'type': Resize, 'args': { 'size': [0.5, 3] } }
+      - EastRandomCropData:
+          size: [960, 960]
+          max_tries: 50
+          keep_ratio: true
+      - MakeBorderMap:
+          shrink_ratio: 0.4
+          thresh_min: 0.3
+          thresh_max: 0.7
+      - MakeShrinkMap:
+          shrink_ratio: 0.4
+          min_text_size: 8
+      - NormalizeImage:
+          scale: 1./255.
+          mean: [0.485, 0.456, 0.406]
+          std: [0.229, 0.224, 0.225]
+          order: 'hwc'
+      - ToCHWImage:
+      - KeepKeys:
+          keep_keys: ['image', 'threshold_map', 'threshold_mask', 'shrink_map', 'shrink_mask'] # the order of the dataloader list
+  loader:
+    shuffle: True
+    drop_last: False
+    batch_size_per_card: 8
+    num_workers: 4
+
+Eval:
+  dataset:
+    name: SimpleDataSet
+    data_dir: ./train_data/icdar2015/text_localization/
+    label_file_list:
+      - ./train_data/icdar2015/text_localization/test_icdar2015_label.txt
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - DetLabelEncode: # Class handling label
+      - DetResizeForTest:
+#           image_shape: [736, 1280]
+      - NormalizeImage:
+          scale: 1./255.
+          mean: [0.485, 0.456, 0.406]
+          std: [0.229, 0.224, 0.225]
+          order: 'hwc'
+      - ToCHWImage:
+      - KeepKeys:
+          keep_keys: ['image', 'shape', 'polys', 'ignore_tags']
+  loader:
+    shuffle: False
+    drop_last: False
+    batch_size_per_card: 1 # must be 1
+    num_workers: 2

configs/e2e/e2e_r50_vd_pg.yml

@@ -7,11 +7,6 @@ Global:
   save_epoch_step: 10
   # evaluation is run every 0 iterationss after the 1000th iteration
   eval_batch_step: [ 0, 1000 ]
-  # 1. If pretrained_model is saved in static mode, such as classification pretrained model
-  #    from static branch, load_static_weights must be set as True.
-  # 2. If you want to finetune the pretrained models we provide in the docs,
-  #    you should set load_static_weights as False.
-  load_static_weights: False
   cal_metric_during_train: False
   pretrained_model:
   checkpoints:
@@ -60,22 +55,25 @@ PostProcess:
   name: PGPostProcess
   score_thresh: 0.5
   mode: fast   # fast or slow two ways
+
 Metric:
   name: E2EMetric
-  gt_mat_dir:    # the dir of gt_mat
+  mode: A   # two ways for eval, A: label from txt,  B: label from gt_mat
+  gt_mat_dir:  ./train_data/total_text/gt  # the dir of gt_mat
   character_dict_path: ppocr/utils/ic15_dict.txt
   main_indicator: f_score_e2e
 
 Train:
   dataset:
     name: PGDataSet
-    label_file_list: [.././train_data/total_text/train/]
+    data_dir: ./train_data/total_text/train
+    label_file_list: [./train_data/total_text/train/train.txt]
     ratio_list: [1.0]
-    data_format: icdar #two data format: icdar/textnet
     transforms:
       - DecodeImage: # load image
           img_mode: BGR
           channel_first: False
+      - E2ELabelEncodeTrain:
       - PGProcessTrain:
           batch_size: 14  # same as loader: batch_size_per_card
           min_crop_size: 24
@@ -92,13 +90,13 @@ Train:
 Eval:
   dataset:
     name: PGDataSet
-    data_dir: ./train_data/
-    label_file_list: [./train_data/total_text/test/]
+    data_dir: ./train_data/total_text/test
+    label_file_list: [./train_data/total_text/test/test.txt]
     transforms:
       - DecodeImage: # load image
           img_mode: RGB
           channel_first: False
-      - E2ELabelEncode:
+      - E2ELabelEncodeTest:
       - E2EResizeForTest:
           max_side_len: 768
       - NormalizeImage:
@@ -108,7 +106,7 @@ Eval:
           order: 'hwc'
       - ToCHWImage:
       - KeepKeys:
-          keep_keys: [ 'image', 'shape', 'polys', 'strs', 'tags', 'img_id']
+          keep_keys: [ 'image', 'shape', 'polys', 'texts', 'ignore_tags', 'img_id']
   loader:
     shuffle: False
     drop_last: False

configs/rec/ch_PP-OCRv2/ch_PP-OCRv2_rec.yml

+Global:
+  debug: false
+  use_gpu: true
+  epoch_num: 800
+  log_smooth_window: 20
+  print_batch_step: 10
+  save_model_dir: ./output/rec_mobile_pp-OCRv2
+  save_epoch_step: 3
+  eval_batch_step: [0, 2000]
+  cal_metric_during_train: true
+  pretrained_model:
+  checkpoints:
+  save_inference_dir:
+  use_visualdl: false
+  infer_img: doc/imgs_words/ch/word_1.jpg
+  character_dict_path: ppocr/utils/ppocr_keys_v1.txt
+  character_type: ch
+  max_text_length: 25
+  infer_mode: false
+  use_space_char: true
+  distributed: true
+  save_res_path: ./output/rec/predicts_mobile_pp-OCRv2.txt
+
+
+Optimizer:
+  name: Adam
+  beta1: 0.9
+  beta2: 0.999
+  lr:
+    name: Piecewise
+    decay_epochs : [700, 800]
+    values : [0.001, 0.0001]
+    warmup_epoch: 5
+  regularizer:
+    name: L2
+    factor: 2.0e-05
+
+
+Architecture:
+  model_type: rec
+  algorithm: CRNN
+  Transform:
+  Backbone:
+    name: MobileNetV1Enhance
+    scale: 0.5
+  Neck:
+    name: SequenceEncoder
+    encoder_type: rnn
+    hidden_size: 64
+  Head:
+    name: CTCHead
+    mid_channels: 96
+    fc_decay: 0.00002
+
+Loss:
+  name: CTCLoss
+
+PostProcess:
+  name: CTCLabelDecode
+
+Metric:
+  name: RecMetric
+  main_indicator: acc
+
+Train:
+  dataset:
+    name: SimpleDataSet
+    data_dir: ./train_data/
+    label_file_list:
+    - ./train_data/train_list.txt
+    transforms:
+    - DecodeImage:
+        img_mode: BGR
+        channel_first: false
+    - RecAug:
+    - CTCLabelEncode:
+    - RecResizeImg:
+        image_shape: [3, 32, 320]
+    - KeepKeys:
+        keep_keys:
+        - image
+        - label
+        - length
+  loader:
+    shuffle: true
+    batch_size_per_card: 128
+    drop_last: true
+    num_workers: 8
+Eval:
+  dataset:
+    name: SimpleDataSet
+    data_dir: ./train_data
+    label_file_list:
+    - ./train_data/val_list.txt
+    transforms:
+    - DecodeImage:
+        img_mode: BGR
+        channel_first: false
+    - CTCLabelEncode:
+    - RecResizeImg:
+        image_shape: [3, 32, 320]
+    - KeepKeys:
+        keep_keys:
+        - image
+        - label
+        - length
+  loader:
+    shuffle: false
+    drop_last: false
+    batch_size_per_card: 128
+    num_workers: 8

configs/rec/ch_PP-OCRv2/ch_PP-OCRv2_rec_distillation.yml

+Global:
+  debug: false
+  use_gpu: true
+  epoch_num: 800
+  log_smooth_window: 20
+  print_batch_step: 10
+  save_model_dir: ./output/rec_pp-OCRv2_distillation
+  save_epoch_step: 3
+  eval_batch_step: [0, 2000]
+  cal_metric_during_train: true
+  pretrained_model:
+  checkpoints:
+  save_inference_dir:
+  use_visualdl: false
+  infer_img: doc/imgs_words/ch/word_1.jpg
+  character_dict_path: ppocr/utils/ppocr_keys_v1.txt
+  character_type: ch
+  max_text_length: 25
+  infer_mode: false
+  use_space_char: true
+  distributed: true
+  save_res_path: ./output/rec/predicts_pp-OCRv2_distillation.txt
+
+
+Optimizer:
+  name: Adam
+  beta1: 0.9
+  beta2: 0.999
+  lr:
+    name: Piecewise
+    decay_epochs : [700, 800]
+    values : [0.001, 0.0001]
+    warmup_epoch: 5
+  regularizer:
+    name: L2
+    factor: 2.0e-05
+
+Architecture:
+  model_type: &model_type "rec"
+  name: DistillationModel
+  algorithm: Distillation
+  Models:
+    Teacher:
+      pretrained:
+      freeze_params: false
+      return_all_feats: true
+      model_type: *model_type
+      algorithm: CRNN
+      Transform:
+      Backbone:
+        name: MobileNetV1Enhance
+        scale: 0.5
+      Neck:
+        name: SequenceEncoder
+        encoder_type: rnn
+        hidden_size: 64
+      Head:
+        name: CTCHead
+        mid_channels: 96
+        fc_decay: 0.00002
+    Student:
+      pretrained:
+      freeze_params: false
+      return_all_feats: true
+      model_type: *model_type
+      algorithm: CRNN
+      Transform:
+      Backbone:
+        name: MobileNetV1Enhance
+        scale: 0.5
+      Neck:
+        name: SequenceEncoder
+        encoder_type: rnn
+        hidden_size: 64
+      Head:
+        name: CTCHead
+        mid_channels: 96
+        fc_decay: 0.00002
+  
+
+Loss:
+  name: CombinedLoss
+  loss_config_list:
+  - DistillationCTCLoss:
+      weight: 1.0
+      model_name_list: ["Student", "Teacher"]
+      key: head_out
+  - DistillationDMLLoss:
+      weight: 1.0
+      act: "softmax"
+      use_log: true
+      model_name_pairs:
+      - ["Student", "Teacher"]
+      key: head_out
+  - DistillationDistanceLoss:
+      weight: 1.0
+      mode: "l2"
+      model_name_pairs:
+      - ["Student", "Teacher"]
+      key: backbone_out
+
+PostProcess:
+  name: DistillationCTCLabelDecode
+  model_name: ["Student", "Teacher"]
+  key: head_out
+
+Metric:
+  name: DistillationMetric
+  base_metric_name: RecMetric
+  main_indicator: acc
+  key: "Student"
+
+Train:
+  dataset:
+    name: SimpleDataSet
+    data_dir: ./train_data/
+    label_file_list:
+    - ./train_data/train_list.txt
+    transforms:
+    - DecodeImage:
+        img_mode: BGR
+        channel_first: false
+    - RecAug:
+    - CTCLabelEncode:
+    - RecResizeImg:
+        image_shape: [3, 32, 320]
+    - KeepKeys:
+        keep_keys:
+        - image
+        - label
+        - length
+  loader:
+    shuffle: true
+    batch_size_per_card: 128
+    drop_last: true
+    num_sections: 1
+    num_workers: 8
+Eval:
+  dataset:
+    name: SimpleDataSet
+    data_dir: ./train_data
+    label_file_list:
+    - ./train_data/val_list.txt
+    transforms:
+    - DecodeImage:
+        img_mode: BGR
+        channel_first: false
+    - CTCLabelEncode:
+    - RecResizeImg:
+        image_shape: [3, 32, 320]
+    - KeepKeys:
+        keep_keys:
+        - image
+        - label
+        - length
+  loader:
+    shuffle: false
+    drop_last: false
+    batch_size_per_card: 128
+    num_workers: 8

configs/rec/ch_PP-OCRv2/ch_PP-OCRv2_rec_enhanced_ctc_loss.yml

+Global:
+  debug: false
+  use_gpu: true
+  epoch_num: 800
+  log_smooth_window: 20
+  print_batch_step: 10
+  save_model_dir: ./output/rec_mobile_pp-OCRv2_enhanced_ctc_loss
+  save_epoch_step: 3
+  eval_batch_step: [0, 2000]
+  cal_metric_during_train: true
+  pretrained_model:
+  checkpoints:
+  save_inference_dir:
+  use_visualdl: false
+  infer_img: doc/imgs_words/ch/word_1.jpg
+  character_dict_path: ppocr/utils/ppocr_keys_v1.txt
+  character_type: ch
+  max_text_length: 25
+  infer_mode: false
+  use_space_char: true
+  distributed: true
+  save_res_path: ./output/rec/predicts_mobile_pp-OCRv2_enhanced_ctc_loss.txt
+
+
+Optimizer:
+  name: Adam
+  beta1: 0.9
+  beta2: 0.999
+  lr:
+    name: Piecewise
+    decay_epochs : [700, 800]
+    values : [0.001, 0.0001]
+    warmup_epoch: 5
+  regularizer:
+    name: L2
+    factor: 2.0e-05
+
+
+Architecture:
+  model_type: rec
+  algorithm: CRNN
+  Transform:
+  Backbone:
+    name: MobileNetV1Enhance
+    scale: 0.5
+  Neck:
+    name: SequenceEncoder
+    encoder_type: rnn
+    hidden_size: 64
+  Head:
+    name: CTCHead
+    mid_channels: 96
+    fc_decay: 0.00002
+    return_feats: true
+
+Loss:
+  name: CombinedLoss
+  loss_config_list:
+  - CTCLoss:
+      use_focal_loss: false
+      weight: 1.0
+  - CenterLoss:
+      weight: 0.05
+      num_classes: 6625
+      feat_dim: 96
+      init_center: false
+      center_file_path: "./train_center.pkl"
+  # you can also try to add ace loss on your own dataset
+  # - ACELoss:
+  #     weight: 0.1
+
+PostProcess:
+  name: CTCLabelDecode
+
+Metric:
+  name: RecMetric
+  main_indicator: acc
+
+Train:
+  dataset:
+    name: SimpleDataSet
+    data_dir: ./train_data/
+    label_file_list:
+    - ./train_data/train_list.txt
+    transforms:
+    - DecodeImage:
+        img_mode: BGR
+        channel_first: false
+    - RecAug:
+    - CTCLabelEncode:
+    - RecResizeImg:
+        image_shape: [3, 32, 320]
+    - KeepKeys:
+        keep_keys:
+        - image
+        - label
+        - length
+        - label_ace
+  loader:
+    shuffle: true
+    batch_size_per_card: 128
+    drop_last: true
+    num_workers: 8
+Eval:
+  dataset:
+    name: SimpleDataSet
+    data_dir: ./train_data
+    label_file_list:
+    - ./train_data/val_list.txt
+    transforms:
+    - DecodeImage:
+        img_mode: BGR
+        channel_first: false
+    - CTCLabelEncode:
+    - RecResizeImg:
+        image_shape: [3, 32, 320]
+    - KeepKeys:
+        keep_keys:
+        - image
+        - label
+        - length
+  loader:
+    shuffle: false
+    drop_last: false
+    batch_size_per_card: 128
+    num_workers: 8

configs/rec/ch_ppocr_v2.0/rec_chinese_common_train_v2.0.yml

@@ -19,6 +19,7 @@ Global:
   max_text_length: 25
   infer_mode: False
   use_space_char: True
+  save_res_path: ./output/rec/predicts_chinese_common_v2.0.txt
 
 
 Optimizer:

configs/rec/ch_ppocr_v2.0/rec_chinese_lite_train_v2.0.yml

@@ -19,6 +19,7 @@ Global:
   max_text_length: 25
   infer_mode: False
   use_space_char: True
+  save_res_path: ./output/rec/predicts_chinese_lite_v2.0.txt
 
 
 Optimizer:

configs/rec/rec_icdar15_train.yml

@@ -10,15 +10,16 @@ Global:
   cal_metric_during_train: True
   pretrained_model:
   checkpoints:
-  save_inference_dir:
+  save_inference_dir: ./
   use_visualdl: False
   infer_img: doc/imgs_words_en/word_10.png
   # for data or label process
-  character_dict_path: ppocr/utils/ic15_dict.txt
-  character_type: ch
+  character_dict_path: ppocr/utils/en_dict.txt
+  character_type: EN
   max_text_length: 25
   infer_mode: False
   use_space_char: False
+  save_res_path: ./output/rec/predicts_ic15.txt
 
 Optimizer:
   name: Adam
@@ -59,8 +60,8 @@ Metric:
 Train:
   dataset:
     name: SimpleDataSet
-    data_dir: ./train_data/
-    label_file_list: ["./train_data/train_list.txt"]
+    data_dir: ./train_data/ic15_data/
+    label_file_list: ["./train_data/ic15_data/rec_gt_train.txt"]
     transforms:
       - DecodeImage: # load image
           img_mode: BGR
@@ -80,8 +81,8 @@ Train:
 Eval:
   dataset:
     name: SimpleDataSet
-    data_dir: ./train_data/
-    label_file_list: ["./train_data/train_list.txt"]
+    data_dir: ./train_data/ic15_data
+    label_file_list: ["./train_data/ic15_data/rec_gt_test.txt"]
     transforms:
       - DecodeImage: # load image
           img_mode: BGR

configs/rec/rec_mtb_nrtr.yml

+Global:
+  use_gpu: True
+  epoch_num: 21
+  log_smooth_window: 20
+  print_batch_step: 10
+  save_model_dir: ./output/rec/nrtr/
+  save_epoch_step: 1
+  # evaluation is run every 2000 iterations
+  eval_batch_step: [0, 2000]
+  cal_metric_during_train: True
+  pretrained_model:
+  checkpoints: 
+  save_inference_dir:
+  use_visualdl: False
+  infer_img: doc/imgs_words_en/word_10.png
+  # for data or label process
+  character_dict_path: 
+  character_type: EN_symbol
+  max_text_length: 25
+  infer_mode: False
+  use_space_char: True
+  save_res_path: ./output/rec/predicts_nrtr.txt
+
+Optimizer:
+  name: Adam
+  beta1: 0.9
+  beta2: 0.99
+  clip_norm: 5.0
+  lr:
+    name: Cosine
+    learning_rate: 0.0005
+    warmup_epoch: 2
+  regularizer:
+    name: 'L2'
+    factor: 0.
+
+Architecture:
+  model_type: rec
+  algorithm: NRTR
+  in_channels: 1
+  Transform:
+  Backbone:
+    name: MTB
+    cnn_num: 2
+  Head:
+    name: Transformer
+    d_model: 512
+    num_encoder_layers: 6
+    beam_size: -1 # When Beam size is greater than 0, it means to use beam search when evaluation.
+    
+
+Loss:
+  name: NRTRLoss
+  smoothing: True
+
+PostProcess:
+  name: NRTRLabelDecode
+
+Metric:
+  name: RecMetric
+  main_indicator: acc
+
+Train:
+  dataset:
+    name: LMDBDataSet
+    data_dir: ./train_data/data_lmdb_release/training/
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - NRTRLabelEncode: # Class handling label
+      - NRTRRecResizeImg:
+          image_shape: [100, 32]
+          resize_type: PIL # PIL or OpenCV
+      - KeepKeys:
+          keep_keys: ['image', 'label', 'length'] # dataloader will return list in this order
+  loader:
+    shuffle: True
+    batch_size_per_card: 512
+    drop_last: True
+    num_workers: 8
+
+Eval:
+  dataset:
+    name: LMDBDataSet
+    data_dir: ./train_data/data_lmdb_release/evaluation/
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - NRTRLabelEncode: # Class handling label
+      - NRTRRecResizeImg:
+          image_shape: [100, 32]
+          resize_type: PIL # PIL or OpenCV
+      - KeepKeys:
+          keep_keys: ['image', 'label', 'length'] # dataloader will return list in this order
+  loader:
+    shuffle: False
+    drop_last: False
+    batch_size_per_card: 256
+    num_workers: 1
+    use_shared_memory: False

configs/rec/rec_mv3_none_bilstm_ctc.yml

@@ -19,6 +19,7 @@ Global:
   max_text_length: 25
   infer_mode: False
   use_space_char: False
+  save_res_path: ./output/rec/predicts_mv3_none_bilstm_ctc.txt
 
 Optimizer:
   name: Adam

configs/rec/rec_mv3_none_none_ctc.yml

@@ -19,6 +19,7 @@ Global:
   max_text_length: 25
   infer_mode: False
   use_space_char: False
+  save_res_path: ./output/rec/predicts_mv3_none_none_ctc.txt
 
 Optimizer:
   name: Adam

configs/rec/rec_mv3_tps_bilstm_att.yml

@@ -19,6 +19,7 @@ Global:
   max_text_length: 25
   infer_mode: False
   use_space_char: False
+  save_res_path: ./output/rec/predicts_mv3_tps_bilstm_att.txt
 
 
 Optimizer:

configs/rec/rec_mv3_tps_bilstm_ctc.yml

@@ -19,6 +19,7 @@ Global:
   max_text_length: 25
   infer_mode: False
   use_space_char: False
+  save_res_path: ./output/rec/predicts_mv3_tps_bilstm_ctc.txt
 
 Optimizer:
   name: Adam

configs/rec/rec_r31_sar.yml

+Global:
+  use_gpu: true
+  epoch_num: 5
+  log_smooth_window: 20
+  print_batch_step: 20
+  save_model_dir: ./sar_rec
+  save_epoch_step: 1
+  # evaluation is run every 2000 iterations
+  eval_batch_step: [0, 2000]
+  cal_metric_during_train: True
+  pretrained_model:
+  checkpoints: 
+  save_inference_dir:
+  use_visualdl: False
+  infer_img: 
+  # for data or label process
+  character_dict_path: ppocr/utils/dict90.txt
+  character_type: EN_symbol
+  max_text_length: 30
+  infer_mode: False
+  use_space_char: False
+  rm_symbol: True
+  save_res_path: ./output/rec/predicts_sar.txt
+
+Optimizer:
+  name: Adam
+  beta1: 0.9
+  beta2: 0.999
+  lr:
+    name: Piecewise
+    decay_epochs: [3, 4]
+    values: [0.001, 0.0001, 0.00001] 
+  regularizer:
+    name: 'L2'
+    factor: 0
+
+Architecture:
+  model_type: rec
+  algorithm: SAR
+  Transform:
+  Backbone:
+    name: ResNet31
+  Head:
+    name: SARHead
+
+Loss:
+  name: SARLoss
+
+PostProcess:
+  name: SARLabelDecode
+
+Metric:
+  name: RecMetric
+
+
+Train:
+  dataset:
+    name: SimpleDataSet
+    label_file_list: ['./train_data/train_list.txt']
+    data_dir: ./train_data/
+    ratio_list: 1.0
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - SARLabelEncode: # Class handling label
+      - SARRecResizeImg:
+          image_shape: [3, 48, 48, 160] # h:48 w:[48,160]
+          width_downsample_ratio: 0.25
+      - KeepKeys:
+          keep_keys: ['image', 'label', 'valid_ratio'] # dataloader will return list in this order
+  loader:
+    shuffle: True
+    batch_size_per_card: 64
+    drop_last: True
+    num_workers: 8
+    use_shared_memory: False
+
+Eval:
+  dataset:
+    name: LMDBDataSet
+    data_dir: ./train_data/data_lmdb_release/evaluation/
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - SARLabelEncode: # Class handling label
+      - SARRecResizeImg:
+          image_shape: [3, 48, 48, 160]
+          width_downsample_ratio: 0.25
+      - KeepKeys:
+          keep_keys: ['image', 'label', 'valid_ratio'] # dataloader will return list in this order
+  loader:
+    shuffle: False
+    drop_last: False
+    batch_size_per_card: 64
+    num_workers: 4
+    use_shared_memory: False
+  

configs/rec/rec_r34_vd_none_bilstm_ctc.yml

@@ -19,6 +19,7 @@ Global:
   max_text_length: 25
   infer_mode: False
   use_space_char: False
+  save_res_path: ./output/rec/predicts_r34_vd_none_bilstm_ctc.txt
 
 Optimizer:
   name: Adam

configs/rec/rec_r34_vd_none_none_ctc.yml

@@ -19,6 +19,7 @@ Global:
   max_text_length: 25
   infer_mode: False
   use_space_char: False
+  save_res_path: ./output/rec/predicts_r34_vd_none_none_ctc.txt
 
 Optimizer:
   name: Adam

configs/rec/rec_r34_vd_tps_bilstm_att.yml

@@ -19,6 +19,7 @@ Global:
   max_text_length: 25
   infer_mode: False
   use_space_char: False
+  save_res_path: ./output/rec/predicts_b3_rare_r34_none_gru.txt
 
 
 Optimizer:

configs/rec/rec_r34_vd_tps_bilstm_ctc.yml

@@ -19,6 +19,7 @@ Global:
   max_text_length: 25
   infer_mode: False
   use_space_char: False
+  save_res_path: ./output/rec/predicts_r34_vd_tps_bilstm_ctc.txt
 
 Optimizer:
   name: Adam
@@ -37,7 +38,7 @@ Architecture:
     name: TPS
     num_fiducial: 20
     loc_lr: 0.1
-    model_name: small
+    model_name: large
   Backbone:
     name: ResNet
     layers: 34

configs/rec/rec_r50_fpn_srn.yml

@@ -20,6 +20,7 @@ Global:
   num_heads: 8
   infer_mode: False
   use_space_char: False
+  save_res_path: ./output/rec/predicts_srn.txt
 
 
 Optimizer:

configs/rec/rec_resnet_stn_bilstm_att.yml

+Global:
+  use_gpu: True
+  epoch_num: 400
+  log_smooth_window: 20
+  print_batch_step: 10
+  save_model_dir: ./output/rec/seed
+  save_epoch_step: 3
+  # evaluation is run every 5000 iterations after the 4000th iteration
+  eval_batch_step: [0, 2000]
+  cal_metric_during_train: True
+  pretrained_model:
+  checkpoints:
+  save_inference_dir:
+  use_visualdl: False
+  infer_img: doc/imgs_words_en/word_10.png
+  # for data or label process
+  character_dict_path: 
+  character_type: EN_symbol
+  max_text_length: 100
+  infer_mode: False
+  use_space_char: False
+  save_res_path: ./output/rec/predicts_seed.txt
+
+
+Optimizer:
+  name: Adadelta
+  weight_deacy: 0.0
+  momentum: 0.9
+  lr:
+    name: Piecewise
+    decay_epochs: [4,5,8]
+    values: [1.0, 0.1, 0.01]
+  regularizer:
+    name: 'L2'
+    factor: 2.0e-05
+
+
+Architecture:
+  model_type: rec
+  algorithm: SEED
+  Transform:
+    name: STN_ON
+    tps_inputsize: [32, 64]
+    tps_outputsize: [32, 100]
+    num_control_points: 20
+    tps_margins: [0.05,0.05]
+    stn_activation: none
+  Backbone:
+    name: ResNet_ASTER
+  Head:
+    name: AsterHead  # AttentionHead
+    sDim: 512
+    attDim: 512
+    max_len_labels: 100
+
+Loss:
+  name: AsterLoss
+
+PostProcess:
+  name: SEEDLabelDecode
+
+Metric:
+  name: RecMetric
+  main_indicator: acc
+  is_filter: True
+
+Train:
+  dataset:
+    name: LMDBDataSet
+    data_dir: ./train_data/data_lmdb_release/training/
+    transforms:
+      - Fasttext:
+          path: "./cc.en.300.bin"
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - SEEDLabelEncode: # Class handling label
+      - RecResizeImg:
+          character_type: en
+          image_shape: [3, 64, 256]
+          padding: False
+      - KeepKeys:
+          keep_keys: ['image', 'label', 'length', 'fast_label'] # dataloader will return list in this order
+  loader:
+    shuffle: True
+    batch_size_per_card: 256
+    drop_last: True
+    num_workers: 6
+
+Eval:
+  dataset:
+    name: LMDBDataSet
+    data_dir: ./train_data/data_lmdb_release/evaluation/
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - SEEDLabelEncode: # Class handling label
+      - RecResizeImg:
+          character_type: en
+          image_shape: [3, 64, 256]
+          padding: False
+      - KeepKeys:
+          keep_keys: ['image', 'label', 'length'] # dataloader will return list in this order
+  loader:
+    shuffle: False
+    drop_last: True
+    batch_size_per_card: 256
+    num_workers: 4

configs/table/table_mv3.yml

+Global:
+  use_gpu: true
+  epoch_num: 50
+  log_smooth_window: 20
+  print_batch_step: 5
+  save_model_dir: ./output/table_mv3/
+  save_epoch_step: 5
+  # evaluation is run every 400 iterations after the 0th iteration
+  eval_batch_step: [0, 400]
+  cal_metric_during_train: True
+  pretrained_model: 
+  checkpoints: 
+  save_inference_dir:
+  use_visualdl: False
+  infer_img: doc/imgs_words/ch/word_1.jpg
+  # for data or label process
+  character_dict_path: ppocr/utils/dict/table_structure_dict.txt
+  character_type: en
+  max_text_length: 100
+  max_elem_length: 500
+  max_cell_num: 500
+  infer_mode: False
+  process_total_num: 0
+  process_cut_num: 0
+
+
+Optimizer:
+  name: Adam
+  beta1: 0.9
+  beta2: 0.999
+  clip_norm: 5.0
+  lr:
+    learning_rate: 0.001
+  regularizer:
+    name: 'L2'
+    factor: 0.00000
+
+Architecture:
+  model_type: table
+  algorithm: TableAttn
+  Backbone:
+    name: MobileNetV3
+    scale: 1.0
+    model_name: small
+    disable_se: True
+  Head:
+    name: TableAttentionHead
+    hidden_size: 256
+    l2_decay: 0.00001
+    loc_type: 2
+
+Loss:
+  name: TableAttentionLoss
+  structure_weight: 100.0
+  loc_weight: 10000.0
+
+PostProcess:
+  name: TableLabelDecode
+
+Metric:
+  name: TableMetric
+  main_indicator: acc
+
+Train:
+  dataset:
+    name: PubTabDataSet
+    data_dir: train_data/table/pubtabnet/train/
+    label_file_path: train_data/table/pubtabnet/PubTabNet_2.0.0_train.jsonl
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - ResizeTableImage:
+          max_len: 488
+      - TableLabelEncode:
+      - NormalizeImage:
+          scale: 1./255.
+          mean: [0.485, 0.456, 0.406]
+          std: [0.229, 0.224, 0.225]
+          order: 'hwc'
+      - PaddingTableImage:
+      - ToCHWImage:
+      - KeepKeys:
+          keep_keys: ['image', 'structure', 'bbox_list', 'sp_tokens', 'bbox_list_mask']
+  loader:
+    shuffle: True
+    batch_size_per_card: 32
+    drop_last: True
+    num_workers: 1
+
+Eval:
+  dataset:
+    name: PubTabDataSet
+    data_dir: train_data/table/pubtabnet/val/
+    label_file_path: train_data/table/pubtabnet/PubTabNet_2.0.0_val.jsonl
+    transforms:
+      - DecodeImage: # load image
+          img_mode: BGR
+          channel_first: False
+      - ResizeTableImage:
+          max_len: 488
+      - TableLabelEncode:
+      - NormalizeImage:
+          scale: 1./255.
+          mean: [0.485, 0.456, 0.406]
+          std: [0.229, 0.224, 0.225]
+          order: 'hwc'
+      - PaddingTableImage:
+      - ToCHWImage:
+      - KeepKeys:
+          keep_keys: ['image', 'structure', 'bbox_list', 'sp_tokens', 'bbox_list_mask']
+  loader:
+    shuffle: False
+    drop_last: False
+    batch_size_per_card: 16
+    num_workers: 1

deploy/android_demo/.gitignore

+*.iml
+.gradle
+/local.properties
+/.idea/*
+.DS_Store
+/build
+/captures
+.externalNativeBuild
+

deploy/android_demo/README.md

+# 如何快速测试
+### 1. 安装最新版本的Android Studio
+可以从 https://developer.android.com/studio 下载。本Demo使用是4.0版本Android Studio编写。
+
+### 2. 按照NDK 20 以上版本
+Demo测试的时候使用的是NDK 20b版本，20版本以上均可以支持编译成功。
+
+如果您是初学者，可以用以下方式安装和测试NDK编译环境。
+点击 File -> New ->New Project，  新建  "Native C++" project
+
+### 3. 导入项目
+点击 File->New->Import Project...， 然后跟着Android Studio的引导导入
+
+
+# 获得更多支持
+前往[端计算模型生成平台EasyEdge](https://ai.baidu.com/easyedge/app/open_source_demo?referrerUrl=paddlelite)，获得更多开发支持：
+
+- Demo APP：可使用手机扫码安装，方便手机端快速体验文字识别
+- SDK：模型被封装为适配不同芯片硬件和操作系统SDK，包括完善的接口，方便进行二次开发

deploy/android_demo/app/.gitignore

+/build

deploy/android_demo/app/build.gradle

+import java.security.MessageDigest
+
+apply plugin: 'com.android.application'
+
+android {
+    compileSdkVersion 29
+    defaultConfig {
+        applicationId "com.baidu.paddle.lite.demo.ocr"
+        minSdkVersion 23
+        targetSdkVersion 29
+        versionCode 1
+        versionName "1.0"
+        testInstrumentationRunner "android.support.test.runner.AndroidJUnitRunner"
+        externalNativeBuild {
+            cmake {
+                cppFlags "-std=c++11 -frtti -fexceptions -Wno-format"
+                arguments '-DANDROID_PLATFORM=android-23', '-DANDROID_STL=c++_shared' ,"-DANDROID_ARM_NEON=TRUE"
+            }
+        }
+        ndk {
+            // abiFilters "arm64-v8a", "armeabi-v7a"
+            abiFilters   "arm64-v8a", "armeabi-v7a"
+            ldLibs "jnigraphics"
+        }
+    }
+    buildTypes {
+        release {
+            minifyEnabled false
+            proguardFiles getDefaultProguardFile('proguard-android-optimize.txt'), 'proguard-rules.pro'
+        }
+    }
+    externalNativeBuild {
+        cmake {
+            path "src/main/cpp/CMakeLists.txt"
+            version "3.10.2"
+        }
+    }
+}
+
+dependencies {
+    implementation fileTree(include: ['*.jar'], dir: 'libs')
+    implementation 'androidx.appcompat:appcompat:1.1.0'
+    implementation 'androidx.constraintlayout:constraintlayout:1.1.3'
+    testImplementation 'junit:junit:4.12'
+    androidTestImplementation 'com.android.support.test:runner:1.0.2'
+    androidTestImplementation 'com.android.support.test.espresso:espresso-core:3.0.2'
+}
+
+def archives = [
+        [
+                'src' : 'https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/paddle_lite_libs_v2_9_0.tar.gz',
+                'dest': 'PaddleLite'
+        ],
+        [
+                'src' : 'https://paddlelite-demo.bj.bcebos.com/libs/android/opencv-4.2.0-android-sdk.tar.gz',
+                'dest': 'OpenCV'
+        ],
+        [
+                'src' : 'https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ocr_v2_for_cpu.tar.gz',
+                'dest' : 'src/main/assets/models'
+        ],
+        [
+                'src' : 'https://paddleocr.bj.bcebos.com/dygraph_v2.0/lite/ch_dict.tar.gz',
+                'dest' : 'src/main/assets/labels'
+        ]
+]
+
+task downloadAndExtractArchives(type: DefaultTask) {
+    doFirst {
+        println "Downloading and extracting archives including libs and models"
+    }
+    doLast {
+        // Prepare cache folder for archives
+        String cachePath = "cache"
+        if (!file("${cachePath}").exists()) {
+            mkdir "${cachePath}"
+        }
+        archives.eachWithIndex { archive, index ->
+            MessageDigest messageDigest = MessageDigest.getInstance('MD5')
+            messageDigest.update(archive.src.bytes)
+            String cacheName = new BigInteger(1, messageDigest.digest()).toString(32)
+            // Download the target archive if not exists
+            boolean copyFiles = !file("${archive.dest}").exists()
+            if (!file("${cachePath}/${cacheName}.tar.gz").exists()) {
+                ant.get(src: archive.src, dest: file("${cachePath}/${cacheName}.tar.gz"))
+                copyFiles = true; // force to copy files from the latest archive files
+            }
+            // Extract the target archive if its dest path does not exists
+            if (copyFiles) {
+                copy {
+                    from tarTree("${cachePath}/${cacheName}.tar.gz")
+                    into "${archive.dest}"
+                }
+            }
+        }
+    }
+}
+preBuild.dependsOn downloadAndExtractArchives
\ No newline at end of file

deploy/android_demo/app/proguard-rules.pro

+# Add project specific ProGuard rules here.
+# You can control the set of applied configuration files using the
+# proguardFiles setting in build.gradle.
+#
+# For more details, see
+#   http://developer.android.com/guide/developing/tools/proguard.html
+
+# If your project uses WebView with JS, uncomment the following
+# and specify the fully qualified class name to the JavaScript interface
+# class:
+#-keepclassmembers class fqcn.of.javascript.interface.for.webview {
+#   public *;
+#}
+
+# Uncomment this to preserve the line number information for
+# debugging stack traces.
+#-keepattributes SourceFile,LineNumberTable
+
+# If you keep the line number information, uncomment this to
+# hide the original source file name.
+#-renamesourcefileattribute SourceFile

deploy/android_demo/app/src/androidTest/java/com/baidu/paddle/lite/demo/ocr/ExampleInstrumentedTest.java

+package com.baidu.paddle.lite.demo.ocr;
+
+import android.content.Context;
+import android.support.test.InstrumentationRegistry;
+import android.support.test.runner.AndroidJUnit4;
+
+import org.junit.Test;
+import org.junit.runner.RunWith;
+
+import static org.junit.Assert.*;
+
+/**
+ * Instrumented test, which will execute on an Android device.
+ *
+ * @see <a href="http://d.android.com/tools/testing">Testing documentation</a>
+ */
+@RunWith(AndroidJUnit4.class)
+public class ExampleInstrumentedTest {
+    @Test
+    public void useAppContext() {
+        // Context of the app under test.
+        Context appContext = InstrumentationRegistry.getTargetContext();
+
+        assertEquals("com.baidu.paddle.lite.demo", appContext.getPackageName());
+    }
+}

deploy/android_demo/app/src/main/AndroidManifest.xml

+<?xml version="1.0" encoding="utf-8"?>
+<manifest xmlns:android="http://schemas.android.com/apk/res/android"
+          package="com.baidu.paddle.lite.demo.ocr">
+
+    <uses-permission android:name="android.permission.WRITE_EXTERNAL_STORAGE"/>
+    <uses-permission android:name="android.permission.READ_EXTERNAL_STORAGE"/>
+    <uses-permission android:name="android.permission.CAMERA"/>
+
+
+    <application
+            android:allowBackup="true"
+            android:icon="@mipmap/ic_launcher"
+            android:label="@string/app_name"
+            android:roundIcon="@mipmap/ic_launcher_round"
+            android:supportsRtl="true"
+            android:theme="@style/AppTheme">
+        <!-- to test MiniActivity, change this to com.baidu.paddle.lite.demo.ocr.MiniActivity -->
+        <activity android:name="com.baidu.paddle.lite.demo.ocr.MainActivity">
+            <intent-filter>
+                <action android:name="android.intent.action.MAIN"/>
+                <category android:name="android.intent.category.LAUNCHER"/>
+            </intent-filter>
+        </activity>
+        <activity
+                android:name="com.baidu.paddle.lite.demo.ocr.SettingsActivity"
+                android:label="Settings">
+        </activity>
+        <provider
+            android:name="androidx.core.content.FileProvider"
+            android:authorities="com.baidu.paddle.lite.demo.ocr.fileprovider"
+            android:exported="false"
+            android:grantUriPermissions="true">
+            <meta-data
+                android:name="android.support.FILE_PROVIDER_PATHS"
+                android:resource="@xml/file_paths"></meta-data>
+        </provider>
+    </application>
+
+</manifest>
\ No newline at end of file

deploy/android_demo/app/src/main/cpp/CMakeLists.txt

+# For more information about using CMake with Android Studio, read the
+# documentation: https://d.android.com/studio/projects/add-native-code.html
+
+# Sets the minimum version of CMake required to build the native library.
+
+cmake_minimum_required(VERSION 3.4.1)
+
+# Creates and names a library, sets it as either STATIC or SHARED, and provides
+# the relative paths to its source code. You can define multiple libraries, and
+# CMake builds them for you. Gradle automatically packages shared libraries with
+# your APK.
+
+set(PaddleLite_DIR "${CMAKE_CURRENT_SOURCE_DIR}/../../../PaddleLite")
+include_directories(${PaddleLite_DIR}/cxx/include)
+
+set(OpenCV_DIR "${CMAKE_CURRENT_SOURCE_DIR}/../../../OpenCV/sdk/native/jni")
+message(STATUS "opencv dir: ${OpenCV_DIR}")
+find_package(OpenCV REQUIRED)
+message(STATUS "OpenCV libraries: ${OpenCV_LIBS}")
+include_directories(${OpenCV_INCLUDE_DIRS})
+aux_source_directory(. SOURCES)
+set(CMAKE_CXX_FLAGS
+        "${CMAKE_CXX_FLAGS} -ffast-math -Ofast -Os"
+        )
+set(CMAKE_CXX_FLAGS
+        "${CMAKE_CXX_FLAGS} -fvisibility=hidden -fvisibility-inlines-hidden -fdata-sections -ffunction-sections"
+        )
+set(CMAKE_SHARED_LINKER_FLAGS
+        "${CMAKE_SHARED_LINKER_FLAGS} -Wl,--gc-sections -Wl,-z,nocopyreloc")
+
+add_library(
+        # Sets the name of the library.
+        Native
+        # Sets the library as a shared library.
+        SHARED
+        # Provides a relative path to your source file(s).
+        ${SOURCES})
+
+find_library(
+        # Sets the name of the path variable.
+        log-lib
+        # Specifies the name of the NDK library that you want CMake to locate.
+        log)
+
+add_library(
+        # Sets the name of the library.
+        paddle_light_api_shared
+        # Sets the library as a shared library.
+        SHARED
+        # Provides a relative path to your source file(s).
+        IMPORTED)
+
+set_target_properties(
+        # Specifies the target library.
+        paddle_light_api_shared
+        # Specifies the parameter you want to define.
+        PROPERTIES
+        IMPORTED_LOCATION
+        ${PaddleLite_DIR}/cxx/libs/${ANDROID_ABI}/libpaddle_light_api_shared.so
+        # Provides the path to the library you want to import.
+)
+
+
+# Specifies libraries CMake should link to your target library. You can link
+# multiple libraries, such as libraries you define in this build script,
+# prebuilt third-party libraries, or system libraries.
+
+target_link_libraries(
+        # Specifies the target library.
+        Native
+        paddle_light_api_shared
+        ${OpenCV_LIBS}
+        GLESv2
+        EGL
+        jnigraphics
+        ${log-lib}
+)
+
+add_custom_command(
+        TARGET Native
+        POST_BUILD
+        COMMAND
+        ${CMAKE_COMMAND} -E copy
+        ${PaddleLite_DIR}/cxx/libs/${ANDROID_ABI}/libc++_shared.so
+        ${CMAKE_LIBRARY_OUTPUT_DIRECTORY}/libc++_shared.so)
+
+add_custom_command(
+        TARGET Native
+        POST_BUILD
+        COMMAND
+        ${CMAKE_COMMAND} -E copy
+        ${PaddleLite_DIR}/cxx/libs/${ANDROID_ABI}/libpaddle_light_api_shared.so
+        ${CMAKE_LIBRARY_OUTPUT_DIRECTORY}/libpaddle_light_api_shared.so)
+
+add_custom_command(
+        TARGET Native
+        POST_BUILD
+        COMMAND
+        ${CMAKE_COMMAND} -E copy
+        ${PaddleLite_DIR}/cxx/libs/${ANDROID_ABI}/libhiai.so
+        ${CMAKE_LIBRARY_OUTPUT_DIRECTORY}/libhiai.so)
+
+add_custom_command(
+        TARGET Native
+        POST_BUILD
+        COMMAND
+        ${CMAKE_COMMAND} -E copy
+        ${PaddleLite_DIR}/cxx/libs/${ANDROID_ABI}/libhiai_ir.so
+        ${CMAKE_LIBRARY_OUTPUT_DIRECTORY}/libhiai_ir.so)
+
+add_custom_command(
+        TARGET Native
+        POST_BUILD
+        COMMAND
+        ${CMAKE_COMMAND} -E copy
+        ${PaddleLite_DIR}/cxx/libs/${ANDROID_ABI}/libhiai_ir_build.so
+        ${CMAKE_LIBRARY_OUTPUT_DIRECTORY}/libhiai_ir_build.so)
\ No newline at end of file

deploy/android_demo/app/src/main/cpp/common.h

+//
+// Created by fu on 4/25/18.
+//
+
+#pragma once
+#import <numeric>
+#import <vector>
+
+#ifdef __ANDROID__
+
+#include <android/log.h>
+
+#define LOG_TAG "OCR_NDK"
+
+#define LOGI(...) __android_log_print(ANDROID_LOG_INFO, LOG_TAG, __VA_ARGS__)
+#define LOGW(...) __android_log_print(ANDROID_LOG_WARN, LOG_TAG, __VA_ARGS__)
+#define LOGE(...) __android_log_print(ANDROID_LOG_ERROR, LOG_TAG, __VA_ARGS__)
+#else
+#include <stdio.h>
+#define LOGI(format, ...)                                                      \
+  fprintf(stdout, "[" LOG_TAG "]" format "\n", ##__VA_ARGS__)
+#define LOGW(format, ...)                                                      \
+  fprintf(stdout, "[" LOG_TAG "]" format "\n", ##__VA_ARGS__)
+#define LOGE(format, ...)                                                      \
+  fprintf(stderr, "[" LOG_TAG "]Error: " format "\n", ##__VA_ARGS__)
+#endif
+
+enum RETURN_CODE { RETURN_OK = 0 };
+
+enum NET_TYPE { NET_OCR = 900100, NET_OCR_INTERNAL = 991008 };
+
+template <typename T> inline T product(const std::vector<T> &vec) {
+  if (vec.empty()) {
+    return 0;
+  }
+  return std::accumulate(vec.begin(), vec.end(), 1, std::multiplies<T>());
+}

deploy/android_demo/app/src/main/cpp/native.cpp

+//
+// Created by fujiayi on 2020/7/5.
+//
+
+#include "native.h"
+#include "ocr_ppredictor.h"
+#include <algorithm>
+#include <paddle_api.h>
+#include <string>
+
+static paddle::lite_api::PowerMode str_to_cpu_mode(const std::string &cpu_mode);
+
+extern "C" JNIEXPORT jlong JNICALL
+Java_com_baidu_paddle_lite_demo_ocr_OCRPredictorNative_init(
+    JNIEnv *env, jobject thiz, jstring j_det_model_path,
+    jstring j_rec_model_path, jstring j_cls_model_path, jint j_thread_num,
+    jstring j_cpu_mode) {
+  std::string det_model_path = jstring_to_cpp_string(env, j_det_model_path);
+  std::string rec_model_path = jstring_to_cpp_string(env, j_rec_model_path);
+  std::string cls_model_path = jstring_to_cpp_string(env, j_cls_model_path);
+  int thread_num = j_thread_num;
+  std::string cpu_mode = jstring_to_cpp_string(env, j_cpu_mode);
+  ppredictor::OCR_Config conf;
+  conf.thread_num = thread_num;
+  conf.mode = str_to_cpu_mode(cpu_mode);
+  ppredictor::OCR_PPredictor *orc_predictor =
+      new ppredictor::OCR_PPredictor{conf};
+  orc_predictor->init_from_file(det_model_path, rec_model_path, cls_model_path);
+  return reinterpret_cast<jlong>(orc_predictor);
+}
+
+/**
+ * "LITE_POWER_HIGH" convert to paddle::lite_api::LITE_POWER_HIGH
+ * @param cpu_mode
+ * @return
+ */
+static paddle::lite_api::PowerMode
+str_to_cpu_mode(const std::string &cpu_mode) {
+  static std::map<std::string, paddle::lite_api::PowerMode> cpu_mode_map{
+      {"LITE_POWER_HIGH", paddle::lite_api::LITE_POWER_HIGH},
+      {"LITE_POWER_LOW", paddle::lite_api::LITE_POWER_HIGH},
+      {"LITE_POWER_FULL", paddle::lite_api::LITE_POWER_FULL},
+      {"LITE_POWER_NO_BIND", paddle::lite_api::LITE_POWER_NO_BIND},
+      {"LITE_POWER_RAND_HIGH", paddle::lite_api::LITE_POWER_RAND_HIGH},
+      {"LITE_POWER_RAND_LOW", paddle::lite_api::LITE_POWER_RAND_LOW}};
+  std::string upper_key;
+  std::transform(cpu_mode.cbegin(), cpu_mode.cend(), upper_key.begin(),
+                 ::toupper);
+  auto index = cpu_mode_map.find(upper_key);
+  if (index == cpu_mode_map.end()) {
+    LOGE("cpu_mode not found %s", upper_key.c_str());
+    return paddle::lite_api::LITE_POWER_HIGH;
+  } else {
+    return index->second;
+  }
+}
+
+extern "C" JNIEXPORT jfloatArray JNICALL
+Java_com_baidu_paddle_lite_demo_ocr_OCRPredictorNative_forward(
+    JNIEnv *env, jobject thiz, jlong java_pointer, jfloatArray buf,
+    jfloatArray ddims, jobject original_image) {
+  LOGI("begin to run native forward");
+  if (java_pointer == 0) {
+    LOGE("JAVA pointer is NULL");
+    return cpp_array_to_jfloatarray(env, nullptr, 0);
+  }
+  cv::Mat origin = bitmap_to_cv_mat(env, original_image);
+  if (origin.size == 0) {
+    LOGE("origin bitmap cannot convert to CV Mat");
+    return cpp_array_to_jfloatarray(env, nullptr, 0);
+  }
+  ppredictor::OCR_PPredictor *ppredictor =
+      (ppredictor::OCR_PPredictor *)java_pointer;
+  std::vector<float> dims_float_arr = jfloatarray_to_float_vector(env, ddims);
+  std::vector<int64_t> dims_arr;
+  dims_arr.resize(dims_float_arr.size());
+  std::copy(dims_float_arr.cbegin(), dims_float_arr.cend(), dims_arr.begin());
+
+  // 这里值有点大，就不调用jfloatarray_to_float_vector了
+  int64_t buf_len = (int64_t)env->GetArrayLength(buf);
+  jfloat *buf_data = env->GetFloatArrayElements(buf, JNI_FALSE);
+  float *data = (jfloat *)buf_data;
+  std::vector<ppredictor::OCRPredictResult> results =
+      ppredictor->infer_ocr(dims_arr, data, buf_len, NET_OCR, origin);
+  LOGI("infer_ocr finished with boxes %ld", results.size());
+  // 这里将std::vector<ppredictor::OCRPredictResult> 序列化成
+  // float数组，传输到java层再反序列化
+  std::vector<float> float_arr;
+  for (const ppredictor::OCRPredictResult &r : results) {
+    float_arr.push_back(r.points.size());
+    float_arr.push_back(r.word_index.size());
+    float_arr.push_back(r.score);
+    for (const std::vector<int> &point : r.points) {
+      float_arr.push_back(point.at(0));
+      float_arr.push_back(point.at(1));
+    }
+    for (int index : r.word_index) {
+      float_arr.push_back(index);
+    }
+  }
+  return cpp_array_to_jfloatarray(env, float_arr.data(), float_arr.size());
+}
+
+extern "C" JNIEXPORT void JNICALL
+Java_com_baidu_paddle_lite_demo_ocr_OCRPredictorNative_release(
+    JNIEnv *env, jobject thiz, jlong java_pointer) {
+  if (java_pointer == 0) {
+    LOGE("JAVA pointer is NULL");
+    return;
+  }
+  ppredictor::OCR_PPredictor *ppredictor =
+      (ppredictor::OCR_PPredictor *)java_pointer;
+  delete ppredictor;
+}
\ No newline at end of file

deploy/android_demo/app/src/main/cpp/native.h

+//
+// Created by fujiayi on 2020/7/5.
+//
+
+#pragma once
+
+#include "common.h"
+#include <android/bitmap.h>
+#include <jni.h>
+#include <opencv2/opencv.hpp>
+#include <string>
+#include <vector>
+
+inline std::string jstring_to_cpp_string(JNIEnv *env, jstring jstr) {
+  // In java, a unicode char will be encoded using 2 bytes (utf16).
+  // so jstring will contain characters utf16. std::string in c++ is
+  // essentially a string of bytes, not characters, so if we want to
+  // pass jstring from JNI to c++, we have convert utf16 to bytes.
+  if (!jstr) {
+    return "";
+  }
+  const jclass stringClass = env->GetObjectClass(jstr);
+  const jmethodID getBytes =
+      env->GetMethodID(stringClass, "getBytes", "(Ljava/lang/String;)[B");
+  const jbyteArray stringJbytes = (jbyteArray)env->CallObjectMethod(
+      jstr, getBytes, env->NewStringUTF("UTF-8"));
+
+  size_t length = (size_t)env->GetArrayLength(stringJbytes);
+  jbyte *pBytes = env->GetByteArrayElements(stringJbytes, NULL);
+
+  std::string ret = std::string(reinterpret_cast<char *>(pBytes), length);
+  env->ReleaseByteArrayElements(stringJbytes, pBytes, JNI_ABORT);
+
+  env->DeleteLocalRef(stringJbytes);
+  env->DeleteLocalRef(stringClass);
+  return ret;
+}
+
+inline jstring cpp_string_to_jstring(JNIEnv *env, std::string str) {
+  auto *data = str.c_str();
+  jclass strClass = env->FindClass("java/lang/String");
+  jmethodID strClassInitMethodID =
+      env->GetMethodID(strClass, "<init>", "([BLjava/lang/String;)V");
+
+  jbyteArray bytes = env->NewByteArray(strlen(data));
+  env->SetByteArrayRegion(bytes, 0, strlen(data),
+                          reinterpret_cast<const jbyte *>(data));
+
+  jstring encoding = env->NewStringUTF("UTF-8");
+  jstring res = (jstring)(
+      env->NewObject(strClass, strClassInitMethodID, bytes, encoding));
+
+  env->DeleteLocalRef(strClass);
+  env->DeleteLocalRef(encoding);
+  env->DeleteLocalRef(bytes);
+
+  return res;
+}
+
+inline jfloatArray cpp_array_to_jfloatarray(JNIEnv *env, const float *buf,
+                                            int64_t len) {
+  if (len == 0) {
+    return env->NewFloatArray(0);
+  }
+  jfloatArray result = env->NewFloatArray(len);
+  env->SetFloatArrayRegion(result, 0, len, buf);
+  return result;
+}
+
+inline jintArray cpp_array_to_jintarray(JNIEnv *env, const int *buf,
+                                        int64_t len) {
+  jintArray result = env->NewIntArray(len);
+  env->SetIntArrayRegion(result, 0, len, buf);
+  return result;
+}
+
+inline jbyteArray cpp_array_to_jbytearray(JNIEnv *env, const int8_t *buf,
+                                          int64_t len) {
+  jbyteArray result = env->NewByteArray(len);
+  env->SetByteArrayRegion(result, 0, len, buf);
+  return result;
+}
+
+inline jlongArray int64_vector_to_jlongarray(JNIEnv *env,
+                                             const std::vector<int64_t> &vec) {
+  jlongArray result = env->NewLongArray(vec.size());
+  jlong *buf = new jlong[vec.size()];
+  for (size_t i = 0; i < vec.size(); ++i) {
+    buf[i] = (jlong)vec[i];
+  }
+  env->SetLongArrayRegion(result, 0, vec.size(), buf);
+  delete[] buf;
+  return result;
+}
+
+inline std::vector<int64_t> jlongarray_to_int64_vector(JNIEnv *env,
+                                                       jlongArray data) {
+  int data_size = env->GetArrayLength(data);
+  jlong *data_ptr = env->GetLongArrayElements(data, nullptr);
+  std::vector<int64_t> data_vec(data_ptr, data_ptr + data_size);
+  env->ReleaseLongArrayElements(data, data_ptr, 0);
+  return data_vec;
+}
+
+inline std::vector<float> jfloatarray_to_float_vector(JNIEnv *env,
+                                                      jfloatArray data) {
+  int data_size = env->GetArrayLength(data);
+  jfloat *data_ptr = env->GetFloatArrayElements(data, nullptr);
+  std::vector<float> data_vec(data_ptr, data_ptr + data_size);
+  env->ReleaseFloatArrayElements(data, data_ptr, 0);
+  return data_vec;
+}
+
+inline cv::Mat bitmap_to_cv_mat(JNIEnv *env, jobject bitmap) {
+  AndroidBitmapInfo info;
+  int result = AndroidBitmap_getInfo(env, bitmap, &info);
+  if (result != ANDROID_BITMAP_RESULT_SUCCESS) {
+    LOGE("AndroidBitmap_getInfo failed, result: %d", result);
+    return cv::Mat{};
+  }
+  if (info.format != ANDROID_BITMAP_FORMAT_RGBA_8888) {
+    LOGE("Bitmap format is not RGBA_8888 !");
+    return cv::Mat{};
+  }
+  unsigned char *srcData = NULL;
+  AndroidBitmap_lockPixels(env, bitmap, (void **)&srcData);
+  cv::Mat mat = cv::Mat::zeros(info.height, info.width, CV_8UC4);
+  memcpy(mat.data, srcData, info.height * info.width * 4);
+  AndroidBitmap_unlockPixels(env, bitmap);
+  cv::cvtColor(mat, mat, cv::COLOR_RGBA2BGR);
+  /**
+  if (!cv::imwrite("/sdcard/1/copy.jpg", mat)){
+      LOGE("Write image failed " );
+  }
+   */
+  return mat;
+}

deploy/cpp_infer/src/clipper.cpp → deploy/android_demo/app/src/main/cpp/ocr_clipper.cpp

@@ -37,6 +37,8 @@
 * used has retained a Delphi flavour.                                          *
 *                                                                              *
 *******************************************************************************/
+
+#include "ocr_clipper.hpp"
 #include <algorithm>
 #include <cmath>
 #include <cstdlib>
@@ -46,8 +48,6 @@
 #include <stdexcept>
 #include <vector>
 
-#include "include/clipper.h"
-
 namespace ClipperLib {
 
 static double const pi = 3.141592653589793238;

deploy/cpp_infer/src/config.cpp → deploy/android_demo/app/src/main/cpp/ocr_cls_process.cpp

@@ -12,53 +12,35 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
-#include <include/config.h>
-
-namespace PaddleOCR {
-
-std::vector<std::string> OCRConfig::split(const std::string &str,
-                                          const std::string &delim) {
-  std::vector<std::string> res;
-  if ("" == str)
-    return res;
-  char *strs = new char[str.length() + 1];
-  std::strcpy(strs, str.c_str());
-
-  char *d = new char[delim.length() + 1];
-  std::strcpy(d, delim.c_str());
-
-  char *p = std::strtok(strs, d);
-  while (p) {
-    std::string s = p;
-    res.push_back(s);
-    p = std::strtok(NULL, d);
+#include "ocr_cls_process.h"
+#include <cmath>
+#include <cstring>
+#include <fstream>
+#include <iostream>
+#include <iostream>
+#include <vector>
+
+const std::vector<int> CLS_IMAGE_SHAPE = {3, 48, 192};
+
+cv::Mat cls_resize_img(const cv::Mat &img) {
+  int imgC = CLS_IMAGE_SHAPE[0];
+  int imgW = CLS_IMAGE_SHAPE[2];
+  int imgH = CLS_IMAGE_SHAPE[1];
+
+  float ratio = float(img.cols) / float(img.rows);
+  int resize_w = 0;
+  if (ceilf(imgH * ratio) > imgW)
+    resize_w = imgW;
+  else
+    resize_w = int(ceilf(imgH * ratio));
+
+  cv::Mat resize_img;
+  cv::resize(img, resize_img, cv::Size(resize_w, imgH), 0.f, 0.f,
+             cv::INTER_CUBIC);
+
+  if (resize_w < imgW) {
+    cv::copyMakeBorder(resize_img, resize_img, 0, 0, 0, int(imgW - resize_w),
+                       cv::BORDER_CONSTANT, {0, 0, 0});
   }
-
-  return res;
-}
-
-std::map<std::string, std::string>
-OCRConfig::LoadConfig(const std::string &config_path) {
-  auto config = Utility::ReadDict(config_path);
-
-  std::map<std::string, std::string> dict;
-  for (int i = 0; i < config.size(); i++) {
-    // pass for empty line or comment
-    if (config[i].size() <= 1 || config[i][0] == '#') {
-      continue;
-    }
-    std::vector<std::string> res = split(config[i], " ");
-    dict[res[0]] = res[1];
-  }
-  return dict;
-}
-
-void OCRConfig::PrintConfigInfo() {
-  std::cout << "=======Paddle OCR inference config======" << std::endl;
-  for (auto iter = config_map_.begin(); iter != config_map_.end(); iter++) {
-    std::cout << iter->first << " : " << iter->second << std::endl;
-  }
-  std::cout << "=======End of Paddle OCR inference config======" << std::endl;
-}
-
-} // namespace PaddleOCR
\ No newline at end of file
+  return resize_img;
+}
\ No newline at end of file

deploy/android_demo/app/src/main/cpp/ocr_cls_process.h

+// Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved.
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+#pragma once
+
+#include "common.h"
+#include <opencv2/opencv.hpp>
+#include <vector>
+
+extern const std::vector<int> CLS_IMAGE_SHAPE;
+
+cv::Mat cls_resize_img(const cv::Mat &img);
\ No newline at end of file

deploy/android_demo/app/src/main/cpp/ocr_crnn_process.cpp

+// Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved.
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+#include "ocr_crnn_process.h"
+#include <cmath>
+#include <cstring>
+#include <fstream>
+#include <iostream>
+#include <iostream>
+#include <vector>
+
+const std::string CHARACTER_TYPE = "ch";
+const int MAX_DICT_LENGTH = 6624;
+const std::vector<int> REC_IMAGE_SHAPE = {3, 32, 320};
+
+static cv::Mat crnn_resize_norm_img(cv::Mat img, float wh_ratio) {
+  int imgC = REC_IMAGE_SHAPE[0];
+  int imgW = REC_IMAGE_SHAPE[2];
+  int imgH = REC_IMAGE_SHAPE[1];
+
+  if (CHARACTER_TYPE == "ch")
+    imgW = int(32 * wh_ratio);
+
+  float ratio = float(img.cols) / float(img.rows);
+  int resize_w = 0;
+  if (ceilf(imgH * ratio) > imgW)
+    resize_w = imgW;
+  else
+    resize_w = int(ceilf(imgH * ratio));
+  cv::Mat resize_img;
+  cv::resize(img, resize_img, cv::Size(resize_w, imgH), 0.f, 0.f,
+             cv::INTER_CUBIC);
+
+  resize_img.convertTo(resize_img, CV_32FC3, 1 / 255.f);
+
+  for (int h = 0; h < resize_img.rows; h++) {
+    for (int w = 0; w < resize_img.cols; w++) {
+      resize_img.at<cv::Vec3f>(h, w)[0] =
+          (resize_img.at<cv::Vec3f>(h, w)[0] - 0.5) * 2;
+      resize_img.at<cv::Vec3f>(h, w)[1] =
+          (resize_img.at<cv::Vec3f>(h, w)[1] - 0.5) * 2;
+      resize_img.at<cv::Vec3f>(h, w)[2] =
+          (resize_img.at<cv::Vec3f>(h, w)[2] - 0.5) * 2;
+    }
+  }
+
+  cv::Mat dist;
+  cv::copyMakeBorder(resize_img, dist, 0, 0, 0, int(imgW - resize_w),
+                     cv::BORDER_CONSTANT, {0, 0, 0});
+
+  return dist;
+}
+
+cv::Mat crnn_resize_img(const cv::Mat &img, float wh_ratio) {
+  int imgC = REC_IMAGE_SHAPE[0];
+  int imgW = REC_IMAGE_SHAPE[2];
+  int imgH = REC_IMAGE_SHAPE[1];
+
+  if (CHARACTER_TYPE == "ch") {
+    imgW = int(32 * wh_ratio);
+  }
+
+  float ratio = float(img.cols) / float(img.rows);
+  int resize_w = 0;
+  if (ceilf(imgH * ratio) > imgW)
+    resize_w = imgW;
+  else
+    resize_w = int(ceilf(imgH * ratio));
+  cv::Mat resize_img;
+  cv::resize(img, resize_img, cv::Size(resize_w, imgH));
+  return resize_img;
+}
+
+cv::Mat get_rotate_crop_image(const cv::Mat &srcimage,
+                              const std::vector<std::vector<int>> &box) {
+
+  std::vector<std::vector<int>> points = box;
+
+  int x_collect[4] = {box[0][0], box[1][0], box[2][0], box[3][0]};
+  int y_collect[4] = {box[0][1], box[1][1], box[2][1], box[3][1]};
+  int left = int(*std::min_element(x_collect, x_collect + 4));
+  int right = int(*std::max_element(x_collect, x_collect + 4));
+  int top = int(*std::min_element(y_collect, y_collect + 4));
+  int bottom = int(*std::max_element(y_collect, y_collect + 4));
+
+  cv::Mat img_crop;
+  srcimage(cv::Rect(left, top, right - left, bottom - top)).copyTo(img_crop);
+
+  for (int i = 0; i < points.size(); i++) {
+    points[i][0] -= left;
+    points[i][1] -= top;
+  }
+
+  int img_crop_width = int(sqrt(pow(points[0][0] - points[1][0], 2) +
+                                pow(points[0][1] - points[1][1], 2)));
+  int img_crop_height = int(sqrt(pow(points[0][0] - points[3][0], 2) +
+                                 pow(points[0][1] - points[3][1], 2)));
+
+  cv::Point2f pts_std[4];
+  pts_std[0] = cv::Point2f(0., 0.);
+  pts_std[1] = cv::Point2f(img_crop_width, 0.);
+  pts_std[2] = cv::Point2f(img_crop_width, img_crop_height);
+  pts_std[3] = cv::Point2f(0.f, img_crop_height);
+
+  cv::Point2f pointsf[4];
+  pointsf[0] = cv::Point2f(points[0][0], points[0][1]);
+  pointsf[1] = cv::Point2f(points[1][0], points[1][1]);
+  pointsf[2] = cv::Point2f(points[2][0], points[2][1]);
+  pointsf[3] = cv::Point2f(points[3][0], points[3][1]);
+
+  cv::Mat M = cv::getPerspectiveTransform(pointsf, pts_std);
+
+  cv::Mat dst_img;
+  cv::warpPerspective(img_crop, dst_img, M,
+                      cv::Size(img_crop_width, img_crop_height),
+                      cv::BORDER_REPLICATE);
+
+  if (float(dst_img.rows) >= float(dst_img.cols) * 1.5) {
+    /*
+    cv::Mat srcCopy = cv::Mat(dst_img.rows, dst_img.cols, dst_img.depth());
+    cv::transpose(dst_img, srcCopy);
+    cv::flip(srcCopy, srcCopy, 0);
+    return srcCopy;
+    */
+    cv::transpose(dst_img, dst_img);
+    cv::flip(dst_img, dst_img, 0);
+    return dst_img;
+  } else {
+    return dst_img;
+  }
+}

deploy/android_demo/app/src/main/cpp/ocr_crnn_process.h

+//
+// Created by fujiayi on 2020/7/3.
+//
+#pragma once
+
+#include "common.h"
+#include <opencv2/opencv.hpp>
+#include <vector>
+
+extern const std::vector<int> REC_IMAGE_SHAPE;
+
+cv::Mat get_rotate_crop_image(const cv::Mat &srcimage,
+                              const std::vector<std::vector<int>> &box);
+
+cv::Mat crnn_resize_img(const cv::Mat &img, float wh_ratio);
+
+template <class ForwardIterator>
+inline size_t argmax(ForwardIterator first, ForwardIterator last) {
+  return std::distance(first, std::max_element(first, last));
+}
\ No newline at end of file

deploy/android_demo/app/src/main/cpp/ocr_db_post_process.h

+//
+// Created by fujiayi on 2020/7/2.
+//
+#pragma once
+#include <opencv2/opencv.hpp>
+#include <vector>
+
+std::vector<std::vector<std::vector<int>>>
+boxes_from_bitmap(const cv::Mat &pred, const cv::Mat &bitmap);
+
+std::vector<std::vector<std::vector<int>>>
+filter_tag_det_res(const std::vector<std::vector<std::vector<int>>> &o_boxes,
+                   float ratio_h, float ratio_w, const cv::Mat &srcimg);
\ No newline at end of file

deploy/android_demo/app/src/main/cpp/ocr_ppredictor.h

+//
+// Created by fujiayi on 2020/7/1.
+//
+
+#pragma once
+
+#include "ppredictor.h"
+#include <opencv2/opencv.hpp>
+#include <paddle_api.h>
+#include <string>
+
+namespace ppredictor {
+
+/**
+ * Config
+ */
+struct OCR_Config {
+  int thread_num = 4; // Thread num
+  paddle::lite_api::PowerMode mode =
+      paddle::lite_api::LITE_POWER_HIGH; // PaddleLite Mode
+};
+
+/**
+ * PolyGone Result
+ */
+struct OCRPredictResult {
+  std::vector<int> word_index;
+  std::vector<std::vector<int>> points;
+  float score;
+};
+
+/**
+ * OCR there are 2 models
+ * 1. First model（det），select polygones to show where are the texts
+ * 2. crop from the origin images, use these polygones to infer
+ */
+class OCR_PPredictor : public PPredictor_Interface {
+public:
+  OCR_PPredictor(const OCR_Config &config);
+
+  virtual ~OCR_PPredictor() {}
+
+  /**
+   * 初始化二个模型的Predictor
+   * @param det_model_content
+   * @param rec_model_content
+   * @return
+   */
+  int init(const std::string &det_model_content,
+           const std::string &rec_model_content,
+           const std::string &cls_model_content);
+  int init_from_file(const std::string &det_model_path,
+                     const std::string &rec_model_path,
+                     const std::string &cls_model_path);
+  /**
+   * Return OCR result
+   * @param dims
+   * @param input_data
+   * @param input_len
+   * @param net_flag
+   * @param origin
+   * @return
+   */
+  virtual std::vector<OCRPredictResult>
+  infer_ocr(const std::vector<int64_t> &dims, const float *input_data,
+            int input_len, int net_flag, cv::Mat &origin);
+
+  virtual NET_TYPE get_net_flag() const;
+
+private:
+  /**
+   * calcul Polygone from the result image of first model
+   * @param pred
+   * @param output_height
+   * @param output_width
+   * @param origin
+   * @return
+   */
+  std::vector<std::vector<std::vector<int>>>
+  calc_filtered_boxes(const float *pred, int pred_size, int output_height,
+                      int output_width, const cv::Mat &origin);
+
+  /**
+   * infer for second model
+   *
+   * @param boxes
+   * @param origin
+   * @return
+   */
+  std::vector<OCRPredictResult>
+  infer_rec(const std::vector<std::vector<std::vector<int>>> &boxes,
+            const cv::Mat &origin);
+
+  /**
+  * infer for cls model
+  *
+  * @param boxes
+  * @param origin
+  * @return
+  */
+  cv::Mat infer_cls(const cv::Mat &origin, float thresh = 0.9);
+
+  /**
+   * Postprocess or sencod model to extract text
+   * @param res
+   * @return
+   */
+  std::vector<int> postprocess_rec_word_index(const PredictorOutput &res);
+
+  /**
+   * calculate confidence of second model text result
+   * @param res
+   * @return
+   */
+  float postprocess_rec_score(const PredictorOutput &res);
+
+  std::unique_ptr<PPredictor> _det_predictor;
+  std::unique_ptr<PPredictor> _rec_predictor;
+  std::unique_ptr<PPredictor> _cls_predictor;
+  OCR_Config _config;
+};
+}

deploy/android_demo/app/src/main/cpp/predictor_input.cpp

+#include "predictor_input.h"
+
+namespace ppredictor {
+
+void PredictorInput::set_dims(std::vector<int64_t> dims) {
+  // yolov3
+  if (_net_flag == 101 && _index == 1) {
+    _tensor->Resize({1, 2});
+    _tensor->mutable_data<int>()[0] = (int)dims.at(2);
+    _tensor->mutable_data<int>()[1] = (int)dims.at(3);
+  } else {
+    _tensor->Resize(dims);
+  }
+  _is_dims_set = true;
+}
+
+float *PredictorInput::get_mutable_float_data() {
+  if (!_is_dims_set) {
+    LOGE("PredictorInput::set_dims is not called");
+  }
+  return _tensor->mutable_data<float>();
+}
+
+void PredictorInput::set_data(const float *input_data, int input_float_len) {
+  float *input_raw_data = get_mutable_float_data();
+  memcpy(input_raw_data, input_data, input_float_len * sizeof(float));
+}
+}
\ No newline at end of file

deploy/android_demo/app/src/main/cpp/preprocess.h

+#pragma once
+
+#include "common.h"
+#include <jni.h>
+#include <opencv2/opencv.hpp>
+cv::Mat bitmap_to_cv_mat(JNIEnv *env, jobject bitmap);
+
+cv::Mat resize_img(const cv::Mat &img, int height, int width);
+
+void neon_mean_scale(const float *din, float *dout, int size,
+                     const std::vector<float> &mean,
+                     const std::vector<float> &scale);