@@ -181,16 +181,11 @@ For a new language request, please refer to [Guideline for new language_requests
...
@@ -181,16 +181,11 @@ For a new language request, please refer to [Guideline for new language_requests
<aname="language_requests"></a>
<aname="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：
If you want to request a new language support, a PR with 1 following files are needed：
1. In folder [ppocr/utils/dict](./ppocr/utils/dict),
1. In folder [ppocr/utils/dict](./ppocr/utils/dict),
it is necessary to submit the dict text to this path and name it with `{language}_dict.txt` that contains a list of all characters. Please see the format example from other files in that folder.
it is necessary to submit the dict text to this path and name it with `{language}_dict.txt` that contains a list of all characters. Please see the format example from other files in that folder.
2. In folder [ppocr/utils/corpus](./ppocr/utils/corpus),
it is necessary to submit the corpus to this path and name it with `{language}_corpus.txt` that contains a list of words in your language.
Maybe, 50000 words per language is necessary at least.
Of course, the more, the better.
If your language has unique elements, please tell me in advance within any way, such as useful links, wikipedia and so on.
If your language has unique elements, please tell me in advance within any way, such as useful links, wikipedia and so on.
More details, please refer to [Multilingual OCR Development Plan](https://github.com/PaddlePaddle/PaddleOCR/issues/1048).
More details, please refer to [Multilingual OCR Development Plan](https://github.com/PaddlePaddle/PaddleOCR/issues/1048).
Generally, a more complex model would achive better performance in the task, but it also leads to some redundancy in the model. Model Pruning is a technique that reduces this redundancy by removing the sub-models in the neural network model, so as to reduce model calculation complexity and improve model inference performance.
Generally, a more complex model would achive better performance in the task, but it also leads to some redundancy in the model. Model Pruning is a technique that reduces this redundancy by removing the sub-models in the neural network model, so as to reduce model calculation complexity and improve model inference performance.
This example uses PaddleSlim provided[APIs of Pruning](https://paddlepaddle.github.io/PaddleSlim/api/prune_api/) to compress the OCR model.
This example uses PaddleSlim provided[APIs of Pruning](https://github.com/PaddlePaddle/PaddleSlim/tree/develop/docs/zh_cn/api_cn/dygraph/pruners) to compress the OCR model.
[PaddleSlim](https://github.com/PaddlePaddle/PaddleSlim), an open source library which integrates model pruning, quantization (including quantization training and offline quantization), distillation, neural network architecture search, and many other commonly used and leading model compression technique in the industry.
[PaddleSlim](https://github.com/PaddlePaddle/PaddleSlim), an open source library which integrates model pruning, quantization (including quantization training and offline quantization), distillation, neural network architecture search, and many other commonly used and leading model compression technique in the industry.
It is recommended that you could understand following pages before reading this example：
It is recommended that you could understand following pages before reading this example：
...
@@ -35,7 +35,7 @@ PaddleOCR also provides a series of [models](../../../doc/doc_en/models_list_en.
...
@@ -35,7 +35,7 @@ PaddleOCR also provides a series of [models](../../../doc/doc_en/models_list_en.
### 3. Pruning sensitivity analysis
### 3. Pruning sensitivity analysis
After the pre-trained model is loaded, sensitivity analysis is performed on each network layer of the model to understand the redundancy of each network layer, and save a sensitivity file which named: sen.pickle. After that, user could load the sensitivity file via the [methods provided by PaddleSlim](https://github.com/PaddlePaddle/PaddleSlim/blob/develop/paddleslim/prune/sensitive.py#L221) and determining the pruning ratio of each network layer automatically. For specific details of sensitivity analysis, see：[Sensitivity analysis](https://github.com/PaddlePaddle/PaddleSlim/blob/develop/docs/zh_cn/tutorials/image_classification_sensitivity_analysis_tutorial.md)
After the pre-trained model is loaded, sensitivity analysis is performed on each network layer of the model to understand the redundancy of each network layer, and save a sensitivity file which named: sen.pickle. After that, user could load the sensitivity file via the [methods provided by PaddleSlim](https://github.com/PaddlePaddle/PaddleSlim/blob/develop/paddleslim/prune/sensitive.py#L221) and determining the pruning ratio of each network layer automatically. For specific details of sensitivity analysis, see：[Sensitivity analysis](https://github.com/PaddlePaddle/PaddleSlim/blob/develop/docs/en/tutorials/image_classification_sensitivity_analysis_tutorial_en.md)
The function would return a dict after loading the sensitivity file. The keys of the dict are name of parameters in each layer. And the value of key is the information about pruning sensitivity of corresponding layer. In example, pruning 10% filter of the layer corresponding to conv10_expand_weights would lead to 0.65% degradation of model performance. The details could be seen at: [Sensitivity analysis](https://github.com/PaddlePaddle/PaddleSlim/blob/develop/docs/zh_cn/algo/algo.md#2-%E5%8D%B7%E7%A7%AF%E6%A0%B8%E5%89%AA%E8%A3%81%E5%8E%9F%E7%90%86)
The function would return a dict after loading the sensitivity file. The keys of the dict are name of parameters in each layer. And the value of key is the information about pruning sensitivity of corresponding layer. In example, pruning 10% filter of the layer corresponding to conv10_expand_weights would lead to 0.65% degradation of model performance. The details could be seen at: [Sensitivity analysis](https://github.com/PaddlePaddle/PaddleSlim/blob/release/2.0-alpha/docs/zh_cn/algo/algo.md)
Enter the PaddleOCR root directory，perform sensitivity analysis on the model with the following command：
Enter the PaddleOCR root directory，perform sensitivity analysis on the model with the following command：
@@ -5,11 +5,11 @@ Generally, a more complex model would achieve better performance in the task, bu
...
@@ -5,11 +5,11 @@ Generally, a more complex model would achieve better performance in the task, bu
Quantization is a technique that reduces this redundancy by reducing the full precision data to a fixed number,
Quantization is a technique that reduces this redundancy by reducing the full precision data to a fixed number,
so as to reduce model calculation complexity and improve model inference performance.
so as to reduce model calculation complexity and improve model inference performance.
This example uses PaddleSlim provided [APIs of Quantization](https://paddlepaddle.github.io/PaddleSlim/api/quantization_api/) to compress the OCR model.
This example uses PaddleSlim provided [APIs of Quantization](https://github.com/PaddlePaddle/PaddleSlim/blob/develop/docs/zh_cn/api_cn/dygraph/quanter/qat.rst) to compress the OCR model.
It is recommended that you could understand following pages before reading this example：
It is recommended that you could understand following pages before reading this example：
-[The training strategy of OCR model](../../../doc/doc_en/quickstart_en.md)
-[The training strategy of OCR model](../../../doc/doc_en/quickstart_en.md)
| model_type | Network Type | rec | Currently support`rec`,`det`,`cls` |
| model_type | Network Type | rec | Currently support`rec`,`det`,`cls` |
| algorithm | Model name | CRNN | See [algorithm_overview](./algorithm_overview_en.md) for the support list |
| algorithm | Model name | CRNN | See [algorithm_overview](./algorithm_overview_en.md) for the support list |
| **Transform** | Set the transformation method | - | Currently only recognition algorithms are supported, see [ppocr/modeling/transform](../../ppocr/modeling/transform) for details |
| **Transform** | Set the transformation method | - | Currently only recognition algorithms are supported, see [ppocr/modeling/transforms](../../ppocr/modeling/transforms) for details |
| name | Transformation class name | TPS | Currently supports `TPS` |
| name | Transformation class name | TPS | Currently supports `TPS` |
| num_fiducial | Number of TPS control points | 20 | Ten on the top and bottom |
| num_fiducial | Number of TPS control points | 20 | Ten on the top and bottom |
@@ -12,25 +12,25 @@ Here we have sorted out some Chinese OCR training and prediction tricks, which a
...
@@ -12,25 +12,25 @@ Here we have sorted out some Chinese OCR training and prediction tricks, which a
At present, ResNet_vd series and MobileNetV3 series are the backbone networks used in PaddleOCR, whether replacing the other backbone networks will help to improve the accuracy? What should be paid attention to when replacing?
At present, ResNet_vd series and MobileNetV3 series are the backbone networks used in PaddleOCR, whether replacing the other backbone networks will help to improve the accuracy? What should be paid attention to when replacing?
-**Tips**
-**Tips**
- Whether text detection or text recognition, the choice of backbone network is a trade-off between prediction effect and prediction efficiency. Generally, a larger backbone network is selected, e.g. ResNet101_vd, then the performance of the detection or recognition is more accurate, but the time cost will increase accordingly. And a smaller backbone network is selected, e.g. MobileNetV3_small_x0_35, the prediction speed is faster, but the accuracy of detection or recognition will be reduced. Fortunately, the detection or recognition effect of different backbone networks is positively correlated with the performance of ImageNet 1000 classification task. [**PaddleClas**](https://github.com/PaddlePaddle/PaddleClas/blob/master/README_en.md) have sorted out the 23 series of classification network structures, such as ResNet_vd、Res2Net、HRNet、MobileNetV3、GhostNet. It provides the top1 accuracy of classification, the time cost of GPU(V100 and T4) and CPU(SD 855), and the 117 pretrained models [**download addresses**](https://paddleclas-en.readthedocs.io/en/latest/models/models_intro_en.html).
- Whether text detection or text recognition, the choice of backbone network is a trade-off between prediction effect and prediction efficiency. Generally, a larger backbone network is selected, e.g. ResNet101_vd, then the performance of the detection or recognition is more accurate, but the time cost will increase accordingly. And a smaller backbone network is selected, e.g. MobileNetV3_small_x0_35, the prediction speed is faster, but the accuracy of detection or recognition will be reduced. Fortunately, the detection or recognition effect of different backbone networks is positively correlated with the performance of ImageNet 1000 classification task. [**PaddleClas**](https://github.com/PaddlePaddle/PaddleClas/blob/release/2.3/docs/en/models/models_intro_en.md) have sorted out the 23 series of classification network structures, such as ResNet_vd、Res2Net、HRNet、MobileNetV3、GhostNet. It provides the top1 accuracy of classification, the time cost of GPU(V100 and T4) and CPU(SD 855), and the 117 pretrained models [**download addresses**](https://paddleclas-en.readthedocs.io/en/latest/models/models_intro_en.html).
- Similar as the 4 stages of ResNet, the replacement of text detection backbone network is to determine those four stages to facilitate the integration of FPN like the object detection heads. In addition, for the text detection problem, the pre trained model in ImageNet1000 can accelerate the convergence and improve the accuracy.
- Similar as the 4 stages of ResNet, the replacement of text detection backbone network is to determine those four stages to facilitate the integration of FPN like the object detection heads. In addition, for the text detection problem, the pre trained model in ImageNet1000 can accelerate the convergence and improve the accuracy.
- In order to replace the backbone network of text recognition, we need to pay attention to the descending position of network width and height stride. Since the ratio between width and height is large in chinese text recognition, the frequency of height decrease is less and the frequency of width decrease is more. You can refer the [modifies of MobileNetV3](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/ppocr/modeling/backbones/rec_mobilenet_v3.py) in PaddleOCR.
- In order to replace the backbone network of text recognition, we need to pay attention to the descending position of network width and height stride. Since the ratio between width and height is large in chinese text recognition, the frequency of height decrease is less and the frequency of width decrease is more. You can refer the [modifies of MobileNetV3](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/ppocr/modeling/backbones/rec_mobilenet_v3.py) in PaddleOCR.
<aname="LongChineseTextRecognition"></a>
<aname="LongChineseTextRecognition"></a>
#### 2、Long Chinese Text Recognition
#### 2、Long Chinese Text Recognition
-**Problem Description**
-**Problem Description**
The maximum resolution of Chinese recognition model during training is [3,32,320], if the text image to be recognized is too long, as shown in the figure below, how to adapt?
The maximum resolution of Chinese recognition model during training is [3,32,320], if the text image to be recognized is too long, as shown in the figure below, how to adapt?
During the training, the training samples are not directly resized to [3,32,320]. At first, the height of samples are resized to 32 and keep the ratio between the width and the height. When the width is less than 320, the excess parts are padding 0. Besides, when the ratio between the width and the height of the samples is larger than 10, these samples will be ignored. When the prediction for one image, do as above, but do not limit the max ratio between the width and the height. When the prediction for an images batch, do as training, but the resized target width is the longest width of the images in the batch. [Code as following](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/tools/infer/predict_rec.py)：
During the training, the training samples are not directly resized to [3,32,320]. At first, the height of samples are resized to 32 and keep the ratio between the width and the height. When the width is less than 320, the excess parts are padding 0. Besides, when the ratio between the width and the height of the samples is larger than 10, these samples will be ignored. When the prediction for one image, do as above, but do not limit the max ratio between the width and the height. When the prediction for an images batch, do as training, but the resized target width is the longest width of the images in the batch. [Code as following](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/tools/infer/predict_rec.py)：
```
```
def resize_norm_img(self, img, max_wh_ratio):
def resize_norm_img(self, img, max_wh_ratio):
imgC, imgH, imgW = self.rec_image_shape
imgC, imgH, imgW = self.rec_image_shape
...
@@ -58,11 +58,11 @@ Here we have sorted out some Chinese OCR training and prediction tricks, which a
...
@@ -58,11 +58,11 @@ Here we have sorted out some Chinese OCR training and prediction tricks, which a
-**Problem Description**
-**Problem Description**
As shown in the figure below, for Chinese and English mixed scenes, in order to facilitate reading and using the recognition results, it is often necessary to recognize the spaces between words. How can this situation be adapted?
As shown in the figure below, for Chinese and English mixed scenes, in order to facilitate reading and using the recognition results, it is often necessary to recognize the spaces between words. How can this situation be adapted?
There are two possible methods for space recognition. (1) Optimize the text detection. For spliting the text at the space in detection results, it needs to divide the text line with space into many segments when label the data for detection. (2) Optimize the text recognition. The space character is introduced into the recognition dictionary. Label the blank line in the training data for text recognition. In addition, we can also concat multiple word lines to synthesize the training data with spaces. PaddleOCR currently uses the second method.
There are two possible methods for space recognition. (1) Optimize the text detection. For spliting the text at the space in detection results, it needs to divide the text line with space into many segments when label the data for detection. (2) Optimize the text recognition. The space character is introduced into the recognition dictionary. Label the blank line in the training data for text recognition. In addition, we can also concat multiple word lines to synthesize the training data with spaces. PaddleOCR currently uses the second method.
