CycleGAN_Tensorlayer

Re-implement CycleGAN in TensorLayer

• Original CycleGAN

• Improved CycleGAN with resize-convolution

Prerequisites:

• TensorLayer

• TensorFlow

• Python

Run:

CUDA_VISIBLE_DEVICES=0 python main.py

(if datasets are collected by yourself, you can use dataset_clean.py or dataset_crop.py to pre-process images)

Theory:

The generator process:

The discriminator process:

Result Improvement

• Data augmentation

• Resize convolution[4]

• Instance normalization[5]

data augmentation:

Instance normalizationcomparision by original paper https://arxiv.org/abs/1607.08022:

Resize convolution (Remove Checkerboard Artifacts):

Final Results:

Reference:

• [1] Original Paper: https://arxiv.org/pdf/1703.10593.pdf

• [2] Original implement in Torch: https://github.com/junyanz/CycleGAN/

• [3] TensorLayer by HaoDong: https://github.com/zsdonghao/tensorlayer

• [4] Resize Convolution: https://distill.pub/2016/deconv-checkerboard/

• [5] Instance Normalization: https://arxiv.org/abs/1607.08022