/ Machine Learning

Simple Tensorflow implementation of StarGAN_v2

Simple Tensorflow implementation of StarGAN_v2

StarGAN

StarGAN v2 — Un-official TensorFlow Implementation.

Overview

overview--2-

Architecture

Generator Mapping Network Style Encoder & Discriminator
g_archi mapping_achi--2- sty_and_dis_archi

Usage

├── dataset
   └── YOUR_DATASET_NAME
       ├── train
           ├── domain1 (domain folder)
               ├── xxx.jpg (domain1 image)
               ├── yyy.png
               ├── ...
           ├── domain2
               ├── aaa.jpg (domain2 image)
               ├── bbb.png
               ├── ...
           ├── domain3
           ├── ...
       ├── test
           ├── zzz.jpg (any content image)
           ├── www.png
           ├── ...

Train

python main.py --dataset celebA-HQ_gender --phase train --batch_size 2 --gpu_num 4

Test

python main.py --dataset celebA-HQ_gender --phase test

Refer test

python main.py --dataset celebA-HQ_gender --phase refer_test --refer_img_path refer_img.jpg

Our results

loss_graph

result1

result2

GitHub

Comments