A BeautyGAN and MobileNet model using web demo based flask, flask-restplus, swagger
BeautyGAN: Instance-level Facial Makeup Transfer with Deep Generative Adversarial Network
MobileNetV2: Computer Vision and PatternRecognition
- Website: https://arxiv.org/abs/1801.04381
Usage
- Python 3.6+
- Tensorflow 1.x
Download pretrained models
- https://pan.baidu.com/s/1wngvgT0qzcKJ5LfLMO7m8A
- https://drive.google.com/drive/folders/1pgVqnF2-rnOxcUQ3SO4JwHUFTdiSe5t9
Save pretrained model, index, checkpoint to models/model_p2
.
+--docs
+--mlib
+--models
| +-- model_dlib
| +-- model_p2 -> make folder, here save models
+--static
...
Swagger main
BeautyGAN main
$ cd flask-deeplearning-service-demo$ pip install -r requirements.txt$ python app.py
or
$ python app_rest.pyIt will deploy :
- Flask app running in http://localhost:8080
- Flask normal mode(optional)
- Flask rest-plus mode with swagger(optional)
- Responsible UI
- Support image drag-and-drop
- Use vanilla JavaScript, HTML and CSS
- RestAPI exception handling
- Dlib facing shape predictor landmark
Image classification demo
Image beautyGAN demo
