AnimeGANv2

「Open Source」. The improved version of AnimeGAN.
「Project Page」 | Landscape photos/videos to anime

News
(2020.12.25) AnimeGANv3 will be released along with its paper in the spring of 2021.
(2021.02.21) The pytorch version of AnimeGANv2 has been released, Be grateful to @bryandlee for his contribution.

Focus:

     Different styles of training have different loss weights!

News:

The improvement directions of AnimeGANv2 mainly include the following 4 points:  

• 1. Solve the problem of high-frequency artifacts in the generated image.

• 2. It is easy to train and directly achieve the effects in the paper.

• 3. Further reduce the number of parameters of the generator network. (generator size: 8.17 Mb), The lite version has a smaller generator model.

• 4. Use new high-quality style data, which come from BD movies as much as possible.

        AnimeGAN can be accessed from here.

Requirements

• python 3.6
• tensorflow-gpu
  • tensorflow-gpu 1.8.0 (ubuntu, GPU 1080Ti or Titan xp, cuda 9.0, cudnn 7.1.3)
  • tensorflow-gpu 1.15.0 (ubuntu, GPU 2080Ti, cuda 10.0.130, cudnn 7.6.0)
• opencv
• numpy
• argparse

Usage

1. Download vgg19

vgg19.npy

2. Download Train/Val Photo dataset

Link

3. Do edge_smooth

python edge_smooth.py --dataset Hayao --img_size 256

4. Calculate the three-channel(BGR) color difference

python data_mean.py --dataset Hayao

5. Train

python main.py --phase train --dataset Hayao --data_mean [13.1360,-8.6698,-4.4661] --epoch 101 --init_epoch 10
For light version: python main.py --phase train --dataset Hayao --data_mean [13.1360,-8.6698,-4.4661] --light --epoch 101 --init_epoch 10

6. Extract the weights of the generator

python get_generator_ckpt.py --checkpoint_dir ../checkpoint/AnimeGAN_Hayao_lsgan_300_300_1_2_10_1 --style_name Hayao

7. Inference

python test.py --checkpoint_dir checkpoint/generator_Hayao_weight --test_dir dataset/test/HR_photo --style_name Hayao/HR_photo

8. Convert video to anime

python video2anime.py --video video/input/お花見.mp4 --checkpoint_dir checkpoint/generator_Paprika_weight

Results

Photo to Paprika Style

Photo to Hayao Style

Photo to Shinkai Style

License

This repo is made freely available to academic and non-academic entities for non-commercial purposes such as academic research, teaching, scientific publications. Permission is granted to use the AnimeGANv2 given that you agree to my license terms. Regarding the request for commercial use, please contact us via email to help you obtain the authorization letter.

Author

Xin Chen