diff --git a/modules/image/Image_gan/style_transfer/Photo2Cartoon/README_en.md b/modules/image/Image_gan/style_transfer/Photo2Cartoon/README_en.md new file mode 100644 index 0000000000000000000000000000000000000000..9d909df96a758dfa1215fb7651e490d0fb5cf11a --- /dev/null +++ b/modules/image/Image_gan/style_transfer/Photo2Cartoon/README_en.md @@ -0,0 +1,95 @@ +# Photo2Cartoon + +|Module Name|Photo2Cartoon| +| :--- | :---: | +|Category|image generation| +|Network|U-GAT-IT| +|Dataset|cartoon_data| +|Fine-tuning supported or not|No| +|Module Size|205MB| +|Latest update date|2021-02-26| +|Data indicators|-| + + +## I.Basic Information + +- ### Application Effect Display + - Sample results： +

+
+

+ + + +- ### Module Introduction + + - This module encapsulates project [photo2cartoon](https://github.com/minivision-ai/photo2cartoon-paddle). + + +## II.Installation + +- ### 1、Environmental Dependence + + - paddlepaddle >= 2.0.0 + + - paddlehub >= 2.0.0 | [How to install PaddleHub]() + +- ### 2、Installation + + - ```shell + $ hub install Photo2Cartoon + ``` + - In case of any problems during installation, please refer to: [Windows_Quickstart]() | [Linux_Quickstart]() | [Mac_Quickstart]() + +## III.Module API Prediction + +- ### 1、Prediction Code Example + + - ```python + import paddlehub as hub + import cv2 + + model = hub.Module(name="Photo2Cartoon") + result = model.Cartoon_GEN(images=[cv2.imread('/PATH/TO/IMAGE')]) + # or + # result = model.Cartoon_GEN(paths=['/PATH/TO/IMAGE']) + ``` + +- ### 2、API + + - ```python + def Cartoon_GEN(images=None, + paths=None, + batch_size=1, + output_dir='output', + visualization=False, + use_gpu=False): + ``` + + - Cartoon style generation API. + + - **Parameters** + + - images (list\[numpy.ndarray\]): image data, ndarray.shape is in the format [H, W, C], BGR; + - paths (list[str]): image path; + - output_dir (str): save path of images; + - batch_size (int): the size of batch; + - visualization (bool): Whether to save the results as picture files; + - use_gpu (bool): use GPU or not; **set the CUDA_VISIBLE_DEVICES environment variable first if you are using GPU** + + **NOTE:** choose one parameter to provide data from paths and images + + - **Return** + - res (list\[numpy.ndarray\]): result list，ndarray.shape is \[H, W, C\] + + + +## IV.Release Note + +* 1.0.0 + + First release + + - ```shell + $ hub install Photo2Cartoon==1.0.0 + ``` diff --git a/modules/image/Image_gan/style_transfer/U2Net_Portrait/README_en.md b/modules/image/Image_gan/style_transfer/U2Net_Portrait/README_en.md new file mode 100644 index 0000000000000000000000000000000000000000..2fc10840371ec96e5d4a8553389af333206a0b8b --- /dev/null +++ b/modules/image/Image_gan/style_transfer/U2Net_Portrait/README_en.md @@ -0,0 +1,102 @@ +# U2Net_Portrait + +|Module Name|U2Net_Portrait| +| :--- | :---: | +|Category|image generation| +|Network|U^2Net| +|Dataset|-| +|Fine-tuning supported or not|No| +|Module Size|254MB| +|Latest update date|2021-02-26| +|Data indicators|-| + + +## I.Basic Information + +- ### Application Effect Display + - Sample results： +

+ +
+ Input image +
+ +
+ Output image +
+

+ + +- ### Module Introduction + + - U2Net_Portrait can be used to create a face portrait. + + +## II.Installation + +- ### 1、Environmental Dependence + + - paddlepaddle >= 2.0.0 + + - paddlehub >= 2.0.0 | [How to install PaddleHub]() + +- ### 2、Installation + + - ```shell + $ hub install U2Net_Portrait + ``` + - In case of any problems during installation, please refer to: [Windows_Quickstart]() | [Linux_Quickstart]() | [Mac_Quickstart]() + +## III.Module API Prediction + +- ### 1、Prediction Code Example + + - ```python + import paddlehub as hub + import cv2 + + model = hub.Module(name="U2Net_Portrait") + result = model.Portrait_GEN(images=[cv2.imread('/PATH/TO/IMAGE')]) + # or + # result = model.Portrait_GEN(paths=['/PATH/TO/IMAGE']) + ``` + +- ### 2、API + + - ```python + def Portrait_GEN(images=None, + paths=None, + scale=1, + batch_size=1, + output_dir='output', + face_detection=True, + visualization=False): + ``` + + - Portrait generation API. + + - **Parameters** + + - images (list\[numpy.ndarray\]): image data, ndarray.shape is in the format [H, W, C], BGR; + - paths (list[str]): image path; + - scale (float) : scale for resizing image；
+ - batch_size (int): the size of batch; + - output_dir (str): save path of images; + - visualization (bool): Whether to save the results as picture files; + + **NOTE:** choose one parameter to provide data from paths and images + + - **Return** + - res (list\[numpy.ndarray\]): result list，ndarray.shape is \[H, W, C\] + + + +## IV.Release Note + +* 1.0.0 + + First release + + - ```shell + $ hub install U2Net_Portrait==1.0.0 + ``` diff --git a/modules/image/Image_gan/style_transfer/UGATIT_100w/README_en.md b/modules/image/Image_gan/style_transfer/UGATIT_100w/README_en.md new file mode 100644 index 0000000000000000000000000000000000000000..6c0606bb7b64026f4c30c50f9fbeeb4dddf298d9 --- /dev/null +++ b/modules/image/Image_gan/style_transfer/UGATIT_100w/README_en.md @@ -0,0 +1,139 @@ +# UGATIT_100w + +|Module Name|UGATIT_100w| +| :--- | :---: | +|Category|image generation| +|Network|U-GAT-IT| +|Dataset|selfie2anime| +|Fine-tuning supported or not|No| +|Module Size|41MB| +|Latest update date|2021-02-26| +|Data indicators|-| + + +## I.Basic Information + +- ### Application Effect Display + - Sample results： +

+ +
+ Input image +
+ +
+ Output image +
+

+ + +- ### Module Introduction + + - UGATIT is a model for style transfer. This module can be used to transfer a face image to cartoon style. For more information, please refer to [UGATIT-Paddle Project](https://github.com/miraiwk/UGATIT-paddle). + + +## II.Installation + +- ### 1、Environmental Dependence + + - paddlepaddle >= 1.8.0 + + - paddlehub >= 1.8.0 | [How to install PaddleHub]() + +- ### 2、Installation + + - ```shell + $ hub install UGATIT_100w + ``` + - In case of any problems during installation, please refer to: [Windows_Quickstart]() | [Linux_Quickstart]() | [Mac_Quickstart]() + +## III.Module API Prediction + +- ### 1、Prediction Code Example + + - ```python + import paddlehub as hub + import cv2 + + model = hub.Module(name="UGATIT_100w") + result = model.style_transfer(images=[cv2.imread('/PATH/TO/IMAGE')]) + # or + # result = model.style_transfer(paths=['/PATH/TO/IMAGE']) + ``` + +- ### 2、API + + - ```python + def style_transfer(images=None, + paths=None, + batch_size=1, + output_dir='output', + visualization=False) + ``` + + - Style transfer API. + + - **Parameters** + + - images (list\[numpy.ndarray\]): image data, ndarray.shape is in the format [H, W, C], BGR; + - paths (list[str]): image path; + - batch_size (int): the size of batch; + - visualization (bool): Whether to save the results as picture files; + - output_dir (str): save path of images; + + **NOTE:** choose one parameter to provide data from paths and images + + - **Return** + - res (list\[numpy.ndarray\]): result list，ndarray.shape 为 \[H, W, C\] + + +## IV.Server Deployment + +- PaddleHub Serving can deploy an online service of style transfer. + +- ### Step 1: Start PaddleHub Serving + + - Run the startup command： + - ```shell + $ hub serving start -m UGATIT_100w + ``` + + - The servitization API is now deployed and the default port number is 8866. + + - **NOTE:** If GPU is used for prediction, set CUDA_VISIBLE_DEVICES environment variable before the service, otherwise it need not be set. + +- ### Step 2: Send a predictive request + + - With a configured server, use the following lines of code to send the prediction request and obtain the result + + - ```python + import requests + import json + import cv2 + import base64 + + + def cv2_to_base64(image): + data = cv2.imencode('.jpg', image)[1] + return base64.b64encode(data.tostring()).decode('utf8') + + # Send an HTTP request + data = {'images':[cv2_to_base64(cv2.imread("/PATH/TO/IMAGE"))]} + headers = {"Content-type": "application/json"} + url = "http://127.0.0.1:8866/predict/UGATIT_100w" + r = requests.post(url=url, headers=headers, data=json.dumps(data)) + + # print prediction results + print(r.json()["results"]) + ``` + + +## V.Release Note + +* 1.0.0 + + First release + + - ```shell + $ hub install UGATIT_100w==1.0.0 + ``` diff --git a/modules/image/Image_gan/style_transfer/animegan_v1_hayao_60/README_en.md b/modules/image/Image_gan/style_transfer/animegan_v1_hayao_60/README_en.md new file mode 100644 index 0000000000000000000000000000000000000000..0fe31afd24aa6b1efd6822b28660f0112bbcacee --- /dev/null +++ b/modules/image/Image_gan/style_transfer/animegan_v1_hayao_60/README_en.md @@ -0,0 +1,149 @@ +# animegan_v1_hayao_60 + +|Module Name|animegan_v1_hayao_60| +| :--- | :---: | +|Category|image generation| +|Network|AnimeGAN| +|Dataset|The Wind Rises| +|Fine-tuning supported or not|No| +|Module Size|18MB| +|Latest update date|2021-07-30| +|Data indicators|-| + + +## I.Basic Information + +- ### Application Effect Display + - Sample results： +

+ +
+ Input Image +
+ +
+ Output Image +
+

+ + + +- ### Module Introduction + + - AnimeGAN V1 is a style transfer model, which can transfer a image style to Miyazaki carton style. For more information, please refer to [AnimeGAN V1 Project](https://github.com/TachibanaYoshino/AnimeGAN). + + +## II.Installation + +- ### 1、Environmental Dependence + + - paddlepaddle >= 1.8.0 + + - paddlehub >= 1.8.0 | [How to install PaddleHub]() + +- ### 2、Installation + + - ```shell + $ hub install animegan_v1_hayao_60 + ``` + - In case of any problems during installation, please refer to: [Windows_Quickstart]() | [Linux_Quickstart]() | [Mac_Quickstart]() + +## III.Module API Prediction + +- ### 1、Prediction Code Example + + - ```python + import paddlehub as hub + import cv2 + + model = hub.Module(name="animegan_v1_hayao_60") + result = model.style_transfer(images=[cv2.imread('/PATH/TO/IMAGE')]) + # or + # result = model.style_transfer(paths=['/PATH/TO/IMAGE']) + ``` + +- ### 2、API + + - ```python + def style_transfer(images=None, + paths=None, + output_dir='output', + visualization=False, + min_size=32, + max_size=1024) + ``` + + - Style transfer API. + + - **Parameters** + + - images (list\[numpy.ndarray\]): image data, ndarray.shape is in the format [H, W, C], BGR; + - paths (list[str]): image path; + - output_dir (str): save path of images; + - visualization (bool): Whether to save the results as picture files; + - min\_size (int): min size of image shape，default is 32； + - max\_size (int): max size of image shape，default is 1024. + + **NOTE:** choose one parameter to provide data from paths and images + + - **Return** + - res (list\[numpy.ndarray\]): result list，ndarray.shape is \[H, W, C\] + + +## IV.Server Deployment + +- PaddleHub Serving can deploy an online service of style transfer. +- ### Step 1: Start PaddleHub Serving + + - Run the startup command： + - ```shell + $ hub serving start -m animegan_v1_hayao_60 + ``` + + - The servitization API is now deployed and the default port number is 8866. + + - **NOTE:** If GPU is used for prediction, set CUDA_VISIBLE_DEVICES environment variable before the service, otherwise it need not be set. + +- ### Step 2: Send a predictive request + + - With a configured server, use the following lines of code to send the prediction request and obtain the result + + - ```python + import requests + import json + import cv2 + import base64 + + + def cv2_to_base64(image): + data = cv2.imencode('.jpg', image)[1] + return base64.b64encode(data.tostring()).decode('utf8') + + # Send an HTTP request + data = {'images':[cv2_to_base64(cv2.imread("/PATH/TO/IMAGE"))]} + headers = {"Content-type": "application/json"} + url = "http://127.0.0.1:8866/predict/animegan_v1_hayao_60" + r = requests.post(url=url, headers=headers, data=json.dumps(data)) + + # print prediction results + print(r.json()["results"]) + ``` + + +## V.Release Note + +* 1.0.0 + + First release + +* 1.0.1 + + Adapt to paddlehub2.0 + +* 1.0.2 + + Delete optional parameter batch_size + + - ```shell + $ hub install animegan_v1_hayao_60==1.0.2 + ``` diff --git a/modules/image/Image_gan/style_transfer/animegan_v2_hayao_64/README_en.md b/modules/image/Image_gan/style_transfer/animegan_v2_hayao_64/README_en.md new file mode 100644 index 0000000000000000000000000000000000000000..ac84636f50bb5b39fa266d3d666e130f4c1c8479 --- /dev/null +++ b/modules/image/Image_gan/style_transfer/animegan_v2_hayao_64/README_en.md @@ -0,0 +1,148 @@ +# animegan_v2_hayao_64 + +|Module Name|animegan_v2_hayao_64| +| :--- | :---: | +|Category|image generation| +|Network|AnimeGAN| +|Dataset|The Wind Rises| +|Fine-tuning supported or not|No| +|Module Size|9.4MB| +|Latest update date|2021-07-30| +|Data indicators|-| + + +## I.Basic Information + +- ### Application Effect Display + - Sample results： +

+ +
+ Input image +
+ +
+ Output image +
+

+ +- ### Module Introduction + + - AnimeGAN V2 is a style transfer model, which can transfer a image style to Miyazaki carton style. For more information, please refer to [AnimeGAN V2 Project](https://github.com/TachibanaYoshino/AnimeGANv2). + + +## II.Installation + +- ### 1、Environmental Dependence + + - paddlepaddle >= 1.8.0 + + - paddlehub >= 1.8.0 | [How to install PaddleHub]() + +- ### 2、Installation + + - ```shell + $ hub install animegan_v2_hayao_64 + ``` + - In case of any problems during installation, please refer to: [Windows_Quickstart]() | [Linux_Quickstart]() | [Mac_Quickstart]() + +## III.Module API Prediction + +- ### 1、Prediction Code Example + + - ```python + import paddlehub as hub + import cv2 + + model = hub.Module(name="animegan_v2_hayao_64") + result = model.style_transfer(images=[cv2.imread('/PATH/TO/IMAGE')]) + # or + # result = model.style_transfer(paths=['/PATH/TO/IMAGE']) + ``` + +- ### 3、API + + - ```python + def style_transfer(images=None, + paths=None, + output_dir='output', + visualization=False, + min_size=32, + max_size=1024) + ``` + + - Style transfer API. + + - **Parameters** + + - images (list\[numpy.ndarray\]): image data, ndarray.shape is in the format [H, W, C], BGR; + - paths (list[str]): image path; + - output_dir (str): save path of images; + - visualization (bool): Whether to save the results as picture files; + - min\_size (int): min size of image shape，default is 32； + - max\_size (int): max size of image shape，default is 1024. + + **NOTE:** choose one parameter to provide data from paths and images + + - **Return** + - res (list\[numpy.ndarray\]): result list，ndarray.shape is \[H, W, C\] + + +## IV.Server Deployment + +- PaddleHub Serving can deploy an online service of style transfer. + +- ### Step 1: Start PaddleHub Serving + + - Run the startup command： + - ```shell + $ hub serving start -m animegan_v2_hayao_64 + ``` + + - The servitization API is now deployed and the default port number is 8866. + + - **NOTE:** If GPU is used for prediction, set CUDA_VISIBLE_DEVICES environment variable before the service, otherwise it need not be set. + +- ### Step 2: Send a predictive request + + - With a configured server, use the following lines of code to send the prediction request and obtain the result + + - ```python + import requests + import json + import cv2 + import base64 + + + def cv2_to_base64(image): + data = cv2.imencode('.jpg', image)[1] + return base64.b64encode(data.tostring()).decode('utf8') + + # Send an HTTP request + data = {'images':[cv2_to_base64(cv2.imread("/PATH/TO/IMAGE"))]} + headers = {"Content-type": "application/json"} + url = "http://127.0.0.1:8866/predict/animegan_v2_hayao_64" + r = requests.post(url=url, headers=headers, data=json.dumps(data)) + + # print prediction results + print(r.json()["results"]) + ``` + + +## V.Release Note + +* 1.0.0 + + First release + +* 1.0.1 + + Adapt to paddlehub2.0 + +* 1.0.2 + + Delete optional parameter batch_size + + - ```shell + $ hub install animegan_v2_hayao_64==1.0.2 + ``` diff --git a/modules/image/Image_gan/style_transfer/animegan_v2_hayao_99/README_en.md b/modules/image/Image_gan/style_transfer/animegan_v2_hayao_99/README_en.md new file mode 100644 index 0000000000000000000000000000000000000000..06d1fba7c5f9222a2589407b096978a1099ebeb7 --- /dev/null +++ b/modules/image/Image_gan/style_transfer/animegan_v2_hayao_99/README_en.md @@ -0,0 +1,148 @@ +# animegan_v2_hayao_99 + +|Module Name|animegan_v2_hayao_99| +| :--- | :---: | +|Category|image generation| +|Network|AnimeGAN| +|Dataset|The Wind Rises| +|Fine-tuning supported or not|No| +|Module Size|9.4MB| +|Latest update date|2021-07-30| +|Data indicators|-| + + +## I.Basic Information + +- ### Application Effect Display + - Sample results： +

+ +
+ Input image +
+ +
+ Output image +
+

+ + +- ### Module Introduction + + - AnimeGAN V2 is a style transfer model, which can transfer a image style to Miyazaki carton style. For more information, please refer to [AnimeGAN V2 Project](https://github.com/TachibanaYoshino/AnimeGANv2). + +## II.Installation + +- ### 1、Environmental Dependence + + - paddlepaddle >= 1.8.0 + + - paddlehub >= 1.8.0 | [How to install PaddleHub]() + +- ### 2、Installation + + - ```shell + $ hub install animegan_v2_hayao_99 + ``` + - In case of any problems during installation, please refer to: [Windows_Quickstart]() | [Linux_Quickstart]() | [Mac_Quickstart]() + +## III.Module API Prediction + + +- ### 1、Prediction Code Example + + - ```python + import paddlehub as hub + import cv2 + + model = hub.Module(name="animegan_v2_hayao_99") + result = model.style_transfer(images=[cv2.imread('/PATH/TO/IMAGE')]) + # or + # result = model.style_transfer(paths=['/PATH/TO/IMAGE']) + ``` + +- ### 2、API + + - ```python + def style_transfer(images=None, + paths=None, + output_dir='output', + visualization=False, + min_size=32, + max_size=1024) + ``` + + - Style transfer API. + + - **Parameters** + + - images (list\[numpy.ndarray\]): image data, ndarray.shape is in the format [H, W, C], BGR; + - paths (list[str]): image path; + - output_dir (str): save path of images; + - visualization (bool): Whether to save the results as picture files; + - min\_size (int): min size of image shape，default is 32； + - max\_size (int): max size of image shape，default is 1024. + + **NOTE:** choose one parameter to provide data from paths and images + + - **Return** + - res (list\[numpy.ndarray\]): result list，ndarray.shape is \[H, W, C\] + + +## IV.Server Deployment + +- PaddleHub Serving can deploy an online service of style transfer. +- ### Step 1: Start PaddleHub Serving + + - Run the startup command： + - ```shell + $ hub serving start -m animegan_v2_hayao_99 + ``` + + - The servitization API is now deployed and the default port number is 8866. + + - **NOTE:** If GPU is used for prediction, set CUDA_VISIBLE_DEVICES environment variable before the service, otherwise it need not be set. + +- ### Step 2: Send a predictive request + + - With a configured server, use the following lines of code to send the prediction request and obtain the result + + - ```python + import requests + import json + import cv2 + import base64 + + + def cv2_to_base64(image): + data = cv2.imencode('.jpg', image)[1] + return base64.b64encode(data.tostring()).decode('utf8') + + # Send an HTTP request + data = {'images':[cv2_to_base64(cv2.imread("/PATH/TO/IMAGE"))]} + headers = {"Content-type": "application/json"} + url = "http://127.0.0.1:8866/predict/animegan_v2_hayao_99" + r = requests.post(url=url, headers=headers, data=json.dumps(data)) + + # print prediction results + print(r.json()["results"]) + ``` + + +## V.Release Note + +* 1.0.0 + + First release + +* 1.0.1 + + Adapt to paddlehub2.0 + +* 1.0.2 + + Delete optional parameter batch_size + + - ```shell + $ hub install animegan_v2_hayao_99==1.0.2 + ``` diff --git a/modules/image/Image_gan/style_transfer/animegan_v2_paprika_74/README_en.md b/modules/image/Image_gan/style_transfer/animegan_v2_paprika_74/README_en.md new file mode 100644 index 0000000000000000000000000000000000000000..4e72e8211c733e4671637afdbf9c2e199780266d --- /dev/null +++ b/modules/image/Image_gan/style_transfer/animegan_v2_paprika_74/README_en.md @@ -0,0 +1,147 @@ +# animegan_v2_paprika_74 + +|Module Name|animegan_v2_paprika_74| +| :--- | :---: | +|Category|image generation| +|Network|AnimeGAN| +|Dataset|Paprika| +|Fine-tuning supported or not|No| +|Module Size|9.4MB| +|Latest update date|2021-02-26| +|Data indicators|-| + + +## I.Basic Information + +- ### Application Effect Display + - Sample results： +

+ +
+ Input Image +
+ +
+ Output Image +
+

+ + +- ### Module Introduction + + - AnimeGAN V2 is a style transfer model, which can transfer a image style to paprika carton style. For more information, please refer to [AnimeGAN V2 Project](https://github.com/TachibanaYoshino/AnimeGANv2). + + +## II.Installation + +- ### 1、Environmental Dependence + + - paddlepaddle >= 1.8.0 + + - paddlehub >= 1.8.0 | [How to install PaddleHub]() + +- ### 2、Installation + + - ```shell + $ hub install animegan_v2_paprika_74 + ``` + - In case of any problems during installation, please refer to: [Windows_Quickstart]() | [Linux_Quickstart]() | [Mac_Quickstart]() + +## III.Module API Prediction + +- ### 1、Prediction Code Example + + - ```python + import paddlehub as hub + import cv2 + + model = hub.Module(name="animegan_v2_paprika_74") + result = model.style_transfer(images=[cv2.imread('/PATH/TO/IMAGE')]) + # or + # result = model.style_transfer(paths=['/PATH/TO/IMAGE']) + ``` + +- ### 2、API + + - ```python + def style_transfer(images=None, + paths=None, + output_dir='output', + visualization=False, + min_size=32, + max_size=1024) + ``` + + - Style transfer API. + + - **Parameters** + + - images (list\[numpy.ndarray\]): image data, ndarray.shape is in the format [H, W, C], BGR; + - paths (list[str]): image path; + - output_dir (str): save path of images; + - visualization (bool): Whether to save the results as picture files; + - min\_size (int): min size of image shape，default is 32； + - max\_size (int): max size of image shape，default is 1024. + + - **Return** + - res (list\[numpy.ndarray\]): result list，ndarray.shape is \[H, W, C\] + + +## IV.Server Deployment + +- PaddleHub Serving can deploy an online service of style transfer. + +- ### Step 1: Start PaddleHub Serving + + - Run the startup command： + - ```shell + $ hub serving start -m animegan_v2_paprika_74 + ``` + + - The servitization API is now deployed and the default port number is 8866. + + - **NOTE:** If GPU is used for prediction, set CUDA_VISIBLE_DEVICES environment variable before the service, otherwise it need not be set. + +- ### Step 2: Send a predictive request + + - With a configured server, use the following lines of code to send the prediction request and obtain the result + + - ```python + import requests + import json + import cv2 + import base64 + + + def cv2_to_base64(image): + data = cv2.imencode('.jpg', image)[1] + return base64.b64encode(data.tostring()).decode('utf8') + + # Send an HTTP request + data = {'images':[cv2_to_base64(cv2.imread("/PATH/TO/IMAGE"))]} + headers = {"Content-type": "application/json"} + url = "http://127.0.0.1:8866/predict/animegan_v2_paprika_74" + r = requests.post(url=url, headers=headers, data=json.dumps(data)) + + # print prediction results + print(r.json()["results"]) + ``` + + +## V.Release Note + +* 1.0.0 + + First release + +* 1.0.1 + + Adapt to paddlehub2.0 + +* 1.0.2 + + Delete optional parameter batch_size + + - ```shell + $ hub install animegan_v2_paprika_74==1.0.2 + ``` diff --git a/modules/image/Image_gan/style_transfer/animegan_v2_paprika_98/README_en.md b/modules/image/Image_gan/style_transfer/animegan_v2_paprika_98/README_en.md new file mode 100644 index 0000000000000000000000000000000000000000..9e16bfb89e81291ea7c81c2df32c4e7681f19441 --- /dev/null +++ b/modules/image/Image_gan/style_transfer/animegan_v2_paprika_98/README_en.md @@ -0,0 +1,149 @@ +# animegan_v2_paprika_98 + +|Module Name|animegan_v2_paprika_98| +| :--- | :---: | +|Category|image generation| +|Network|AnimeGAN| +|Dataset|Paprika| +|Fine-tuning supported or not|No| +|Module Size|9.4MB| +|Latest update date|2021-07-30| +|Data indicators|-| + + +## I.Basic Information + +- ### Application Effect Display + - Sample results： +

+ +
+ Input image +
+ +
+ Output image +
+

+ + +- ### Module Introduction + + - AnimeGAN V2 is a style transfer model, which can transfer a image style to paprika carton style. For more information, please refer to [AnimeGAN V2 Project](https://github.com/TachibanaYoshino/AnimeGANv2). + + +## II.Installation + +- ### 1、Environmental Dependence + + - paddlepaddle >= 1.8.0 + + - paddlehub >= 1.8.0 | [How to install PaddleHub]() + +- ### 2、Installation + + - ```shell + $ hub install animegan_v2_paprika_98 + ``` + - In case of any problems during installation, please refer to: [Windows_Quickstart]() | [Linux_Quickstart]() | [Mac_Quickstart]() + +## III.Module API Prediction + +- ### 1、Prediction Code Example + + - ```python + import paddlehub as hub + import cv2 + + model = hub.Module(name="animegan_v2_paprika_98") + result = model.style_transfer(images=[cv2.imread('/PATH/TO/IMAGE')]) + # or + # result = model.style_transfer(paths=['/PATH/TO/IMAGE']) + ``` + +- ### 2、API + + - ```python + def style_transfer(images=None, + paths=None, + output_dir='output', + visualization=False, + min_size=32, + max_size=1024) + ``` + + - Style transfer API. + + - **Parameters** + + - images (list\[numpy.ndarray\]): image data, ndarray.shape is in the format [H, W, C], BGR; + - paths (list[str]): image path; + - output_dir (str): save path of images; + - visualization (bool): Whether to save the results as picture files; + - min\_size (int): min size of image shape，default is 32； + - max\_size (int): max size of image shape，default is 1024. + + **NOTE:** choose one parameter to provide data from paths and images + + - **Return** + - res (list\[numpy.ndarray\]): result list，ndarray.shape is \[H, W, C\] + + +## IV.Server Deployment + +- PaddleHub Serving can deploy an online service of style transfer. + +- ### Step 1: Start PaddleHub Serving + + - Run the startup command： + - ```shell + $ hub serving start -m animegan_v2_paprika_98 + ``` + + - The servitization API is now deployed and the default port number is 8866. + + - **NOTE:** If GPU is used for prediction, set CUDA_VISIBLE_DEVICES environment variable before the service, otherwise it need not be set. + +- ### Step 2: Send a predictive request + + - With a configured server, use the following lines of code to send the prediction request and obtain the result + + - ```python + import requests + import json + import cv2 + import base64 + + + def cv2_to_base64(image): + data = cv2.imencode('.jpg', image)[1] + return base64.b64encode(data.tostring()).decode('utf8') + + # Send an HTTP request + data = {'images':[cv2_to_base64(cv2.imread("/PATH/TO/IMAGE"))]} + headers = {"Content-type": "application/json"} + url = "http://127.0.0.1:8866/predict/animegan_v2_paprika_98" + r = requests.post(url=url, headers=headers, data=json.dumps(data)) + + # print prediction results + print(r.json()["results"]) + ``` + + +## V.Release Note + +* 1.0.0 + + First release + +* 1.0.1 + + Adapt to paddlehub2.0 + +* 1.0.2 + + Delete optional parameter batch_size + + - ```shell + $ hub install animegan_v2_paprika_98==1.0.2 + ``` diff --git a/modules/image/Image_gan/style_transfer/animegan_v2_shinkai_33/README_en.md b/modules/image/Image_gan/style_transfer/animegan_v2_shinkai_33/README_en.md new file mode 100644 index 0000000000000000000000000000000000000000..383d5813a5dd33069dd0f8a6317a85469dbf4221 --- /dev/null +++ b/modules/image/Image_gan/style_transfer/animegan_v2_shinkai_33/README_en.md @@ -0,0 +1,150 @@ +# animegan_v2_shinkai_33 + +|Module Name|animegan_v2_shinkai_33| +| :--- | :---: | +|Category|image generation| +|Network|AnimeGAN| +|Dataset|Your Name, Weathering with you| +|Fine-tuning supported or not|No| +|Module Size|9.4MB| +|Latest update date|2021-07-30| +|Data indicators|-| + + +## I.Basic Information + +- ### Application Effect Display + - Sample results： +

+ +
+ Input image +
+ +
+ Output image +
+

+ + +- ### Module Introduction + + - AnimeGAN V2 is a style transfer model, which can transfer a image style to Shinkai carton style. For more information, please refer to [AnimeGAN V2 Project](https://github.com/TachibanaYoshino/AnimeGANv2). + + +## II.Installation + +- ### 1、Environmental Dependence + + - paddlepaddle >= 1.8.0 + + - paddlehub >= 1.8.0 | [How to install PaddleHub]() + +- ### 2、Installation + + - ```shell + $ hub install animegan_v2_shinkai_33 + ``` + - In case of any problems during installation, please refer to: [Windows_Quickstart]() | [Linux_Quickstart]() | [Mac_Quickstart]() + +## III.Module API Prediction + +- ### 1、Prediction Code Example + + - ```python + import paddlehub as hub + import cv2 + + model = hub.Module(name="animegan_v2_shinkai_33") + result = model.style_transfer(images=[cv2.imread('/PATH/TO/IMAGE')]) + # or + # result = model.style_transfer(paths=['/PATH/TO/IMAGE']) + ``` + +- ### 2、API + + - ```python + def style_transfer(images=None, + paths=None, + output_dir='output', + visualization=False, + min_size=32, + max_size=1024) + ``` + + - Style transfer API. + + - **Parameters** + + - images (list\[numpy.ndarray\]): image data, ndarray.shape is in the format [H, W, C], BGR; + - paths (list[str]): image path; + - output_dir (str): save path of images; + - visualization (bool): Whether to save the results as picture files; + - min\_size (int): min size of image shape，default is 32； + - max\_size (int): max size of image shape，default is 1024. + + **NOTE:** choose one parameter to provide data from paths and images + + - **Return** + - res (list\[numpy.ndarray\]): result list，ndarray.shape is \[H, W, C\] + + +## IV.Server Deployment + +- PaddleHub Serving can deploy an online service of style transfer. + + +- ### Step 1: Start PaddleHub Serving + + - Run the startup command： + - ```shell + $ hub serving start -m animegan_v2_shinkai_33 + ``` + + - The servitization API is now deployed and the default port number is 8866. + + - **NOTE:** If GPU is used for prediction, set CUDA_VISIBLE_DEVICES environment variable before the service, otherwise it need not be set. + +- ### Step 2: Send a predictive request + + - With a configured server, use the following lines of code to send the prediction request and obtain the result + + - ```python + import requests + import json + import cv2 + import base64 + + + def cv2_to_base64(image): + data = cv2.imencode('.jpg', image)[1] + return base64.b64encode(data.tostring()).decode('utf8') + + # Send an HTTP request + data = {'images':[cv2_to_base64(cv2.imread("/PATH/TO/IMAGE"))]} + headers = {"Content-type": "application/json"} + url = "http://127.0.0.1:8866/predict/animegan_v2_shinkai_33" + r = requests.post(url=url, headers=headers, data=json.dumps(data)) + + # print prediction results + print(r.json()["results"]) + ``` + + +## V.Release Note + +* 1.0.0 + + First release + +* 1.0.1 + + Adapt to paddlehub2.0 + +* 1.0.2 + + Delete optional parameter batch_size + + - ```shell + $ hub install animegan_v2_shinkai_33==1.0.2 + ``` diff --git a/modules/image/Image_gan/style_transfer/animegan_v2_shinkai_53/README_en.md b/modules/image/Image_gan/style_transfer/animegan_v2_shinkai_53/README_en.md new file mode 100644 index 0000000000000000000000000000000000000000..3ab8c6b6913a002f3604c3b70c13bb15bbcbab2a --- /dev/null +++ b/modules/image/Image_gan/style_transfer/animegan_v2_shinkai_53/README_en.md @@ -0,0 +1,149 @@ +# animegan_v2_shinkai_53 + +|Module Name|animegan_v2_shinkai_53| +| :--- | :---: | +|Category|image generation| +|Network|AnimeGAN| +|Dataset|Your Name, Weathering with you| +|Fine-tuning supported or not|No| +|Module Size|9.4MB| +|Latest update date|2021-07-30| +|Data indicators|-| + + +## I.Basic Information + +- ### Application Effect Display + - Sample results： +

+ +
+ Input image +
+ +
+ Output image +
+

+ + +- ### Module Introduction + + - AnimeGAN V2 is a style transfer model, which can transfer a image style to Shinkai carton style. For more information, please refer to [AnimeGAN V2 Project](https://github.com/TachibanaYoshino/AnimeGANv2). + + +## II.Installation + +- ### 1、Environmental Dependence + + - paddlepaddle >= 1.8.0 + + - paddlehub >= 1.8.0 | [How to install PaddleHub]() + +- ### 2、Installation + + - ```shell + $ hub install animegan_v2_shinkai_53 + ``` + - In case of any problems during installation, please refer to: [Windows_Quickstart]() | [Linux_Quickstart]() | [Mac_Quickstart]() + +## III.Module API Prediction + +- ### 1、Prediction Code Example + + - ```python + import paddlehub as hub + import cv2 + + model = hub.Module(name="animegan_v2_shinkai_53") + result = model.style_transfer(images=[cv2.imread('/PATH/TO/IMAGE')]) + # or + # result = model.style_transfer(paths=['/PATH/TO/IMAGE']) + ``` + +- ### 2、API + + - ```python + def style_transfer(images=None, + paths=None, + output_dir='output', + visualization=False, + min_size=32, + max_size=1024) + ``` + + - Style transfer API. + + - **Parameters** + + - images (list\[numpy.ndarray\]): image data, ndarray.shape is in the format [H, W, C], BGR; + - paths (list[str]): image path; + - output_dir (str): save path of images; + - visualization (bool): Whether to save the results as picture files; + - min\_size (int): min size of image shape，default is 32； + - max\_size (int): max size of image shape，default is 1024. + + **NOTE:** choose one parameter to provide data from paths and images + + - **Return** + - res (list\[numpy.ndarray\]): result list，ndarray.shape is \[H, W, C\] + + +## IV.Server Deployment + +- PaddleHub Serving can deploy an online service of style transfer. + +- ### Step 1: Start PaddleHub Serving + + - Run the startup command： + - ```shell + $ hub serving start -m animegan_v2_shinkai_53 + ``` + + - The servitization API is now deployed and the default port number is 8866. + + - **NOTE:** If GPU is used for prediction, set CUDA_VISIBLE_DEVICES environment variable before the service, otherwise it need not be set. + +- ### Step 2: Send a predictive request + + - With a configured server, use the following lines of code to send the prediction request and obtain the result + + - ```python + import requests + import json + import cv2 + import base64 + + + def cv2_to_base64(image): + data = cv2.imencode('.jpg', image)[1] + return base64.b64encode(data.tostring()).decode('utf8') + + # Send an HTTP request + data = {'images':[cv2_to_base64(cv2.imread("/PATH/TO/IMAGE"))]} + headers = {"Content-type": "application/json"} + url = "http://127.0.0.1:8866/predict/animegan_v2_shinkai_53" + r = requests.post(url=url, headers=headers, data=json.dumps(data)) + + # print prediction results + print(r.json()["results"]) + ``` + + +## V.Release Note + +* 1.0.0 + + First release + +* 1.0.1 + + Adapt to paddlehub2.0 + +* 1.0.2 + + Delete optional parameter batch_size + + - ```shell + $ hub install animegan_v2_shinkai_53==1.0.2 + ``` diff --git a/modules/image/Image_gan/style_transfer/stylepro_artistic/README_en.md b/modules/image/Image_gan/style_transfer/stylepro_artistic/README_en.md new file mode 100644 index 0000000000000000000000000000000000000000..355c32aeb5098a7bb7425201d97e883b4d204a56 --- /dev/null +++ b/modules/image/Image_gan/style_transfer/stylepro_artistic/README_en.md @@ -0,0 +1,186 @@ +# stylepro_artistic + +|Module Name|stylepro_artistic| +| :--- | :---: | +|Category|image generation| +|Network|StyleProNet| +|Dataset|MS-COCO + WikiArt| +|Fine-tuning supported or not|No| +|Module Size|28MB| +|Latest update date|2021-02-26| +|Data indicators|-| + + +## I.Basic Information + +- ### Application Effect Display + - Sample results： +

+
+

+ +- ### Module Introduction + + - StyleProNet is a model for style transfer, which is light-weight and responds quickly. This module is based on StyleProNet, trained on WikiArt(MS-COCO) and WikiArt(style) datasets, and can be used for style transfer. For more information, please refer to [StyleProNet](https://arxiv.org/abs/2003.07694). + + +## II.Installation + +- ### 1、Environmental Dependence + + - paddlepaddle >= 1.6.2 + + - paddlehub >= 1.6.0 | [How to install PaddleHub]() + +- ### 2、Installation + + - ```shell + $ hub install stylepro_artistic + ``` + - In case of any problems during installation, please refer to: [Windows_Quickstart]() | [Linux_Quickstart]() | [Mac_Quickstart]() + +## III.Module API Prediction + +- ### 1、Command line Prediction + + - ```shell + $ hub run stylepro_artistic --input_path "/PATH/TO/IMAGE" + ``` + - If you want to call the Hub module through the command line, please refer to: [PaddleHub Command Line Instruction](../../../../docs/docs_ch/tutorial/cmd_usage.rst) +- ### 2、Prediction Code Example + + - ```python + import paddlehub as hub + import cv2 + + stylepro_artistic = hub.Module(name="stylepro_artistic") + result = stylepro_artistic.style_transfer( + images=[{ + 'content': cv2.imread('/PATH/TO/CONTENT_IMAGE'), + 'styles': [cv2.imread('/PATH/TO/STYLE_IMAGE')] + }]) + + # or + # result = stylepro_artistic.style_transfer( + # paths=[{ + # 'content': '/PATH/TO/CONTENT_IMAGE', + # 'styles': ['/PATH/TO/STYLE_IMAGE'] + # }]) + ``` + +- ### 3、API + + - ```python + def style_transfer(images=None, + paths=None, + alpha=1, + use_gpu=False, + visualization=False, + output_dir='transfer_result') + ``` + + - Style transfer API. + + - **Parameters** + - images (list\[dict\]): each element is a dict，includes: + - content (numpy.ndarray): input image array，shape is \[H, W, C\]，BGR format；
+ - styles (list\[numpy.ndarray\]) : list of style image arrays，shape is \[H, W, C\]，BGR format；
+ - weights (list\[float\], optioal) : weight for each style, if not set, each style has the same weight;
+ - paths (list\[dict\]): each element is a dict，includes: + - content (str): path for input image；
+ - styles (list\[str\]) : paths for style images；
+ - weights (list\[float\], optioal) : weight for each style, if not set, each style has the same weight;
+ - alpha (float) : alpha value，\[0, 1\] ，default is 1
+ - use_gpu (bool): use GPU or not; **set the CUDA_VISIBLE_DEVICES environment variable first if you are using GPU** + - visualization (bool): Whether to save the results as picture files; + - output_dir (str): save path of images; + + **NOTE:** choose one parameter to provide data from paths and images + + - **Return** + + - res (list\[dict\]): results + - path (str): path for input image + - data (numpy.ndarray): output image + + + - ```python + def save_inference_model(dirname, + model_filename=None, + params_filename=None, + combined=True) + ``` + - Save model to specific path + + - **Parameters** + + - dirname: output dir for saving model + - model\_filename: filename for saving model + - params\_filename: filename for saving parameters + - combined: whether save parameters into one file + + +## IV.Server Deployment + +- PaddleHub Serving can deploy an online service of style transfer. + +- ### Step 1: Start PaddleHub Serving + + - Run the startup command： + - ```shell + $ hub serving start -m stylepro_artistic + ``` + + - The servitization API is now deployed and the default port number is 8866. + + - **NOTE:** If GPU is used for prediction, set CUDA_VISIBLE_DEVICES environment variable before the service, otherwise it need not be set. + +- ### Step 2: Send a predictive request + + - With a configured server, use the following lines of code to send the prediction request and obtain the result + + - ```python + import requests + import json + import cv2 + import base64 + import numpy as np + + + def cv2_to_base64(image): + data = cv2.imencode('.jpg', image)[1] + return base64.b64encode(data.tostring()).decode('utf8') + + def base64_to_cv2(b64str): + data = base64.b64decode(b64str.encode('utf8')) + data = np.fromstring(data, np.uint8) + data = cv2.imdecode(data, cv2.IMREAD_COLOR) + return data + + # Send an HTTP request + data = {'images':[ + { + 'content':cv2_to_base64(cv2.imread('/PATH/TO/CONTENT_IMAGE')), + 'styles':[cv2_to_base64(cv2.imread('/PATH/TO/STYLE_IMAGE'))] + } + ]} + headers = {"Content-type": "application/json"} + url = "http://127.0.0.1:8866/predict/stylepro_artistic" + r = requests.post(url=url, headers=headers, data=json.dumps(data)) + + # print prediction results + print(base64_to_cv2(r.json()["results"][0]['data'])) + ``` + + +## V.Release Note + +* 1.0.0 + + First release + +* 1.0.1 + + - ```shell + $ hub install stylepro_artistic==1.0.1 + ```