/ Machine Learning

Simple Tensorflow implementation of FusionGAN

Simple Tensorflow implementation of FusionGAN

FusionGAN-Tensorflow

Simple Tensorflow implementation of FusionGAN (CVPR 2018).

Requirements

  • Tensorflow 1.8
  • Python 3.6

Usage

├── dataset
   └── YOUR_DATASET_NAME
       ├── trainA
           ├── xxx.jpg (name, format doesn't matter)
           ├── yyy.png
           └── ...
       ├── trainB
           ├── zzz.jpg
           ├── www.png
           └── ...
       ├── testA
           ├── aaa.jpg 
           ├── bbb.png
           └── ...
       └── testB
           ├── ccc.jpg 
           ├── ddd.png
           └── ...

3. Train

  • python main.py --phase train --dataset pose --epoch 20

4. Test

  • python main.py --phase test --dataset pose --epoch 20

Summary

summary

Architecture

architecture

Min-Pooling

minpooling

Shape loss

shape_loss

Results

Our results

result

Paper results

p_result2

p_result

p_result3

GitHub