XiaoheJu / pytorch-CycleGAN

Pytorch implementation of CycleGAN.

pytorch-CycleGAN

Pytorch implementation of CycleGAN [1].

• you can download datasets: https://people.eecs.berkeley.edu/~taesung_park/CycleGAN/datasets/
• you can see more information for network architecture and training details in https://arxiv.org/pdf/1703.10593.pdf

dataset

• apple2orange
  • apple training images: 995, orange training images: 1,019, apple test images: 266, orange test images: 248
• horse2zebra
  • horse training images: 1,067, zebra training images: 1,334, horse test images: 120, zebra test images: 140

Resutls

apple2orange (after 200 epochs)

• apple2orange

Input	Output	Reconstruction

• orange2apple

Input	Output	Reconstruction

• Learning Time
  • apple2orange - Avg. per epoch: 299.38 sec; Total 200 epochs: 62,225.33 sec

horse2zebra (after 200 epochs)

• horse2zebra

Input	Output	Reconstruction

• zebra2horse

Input	Output	Reconstruction

• Learning Time
  • horse2zebra - Avg. per epoch: 299.25 sec; Total 200 epochs: 61,221.27 sec

Development Environment

• Ubuntu 14.04 LTS
• NVIDIA GTX 1080 ti
• cuda 8.0
• Python 2.7.6
• pytorch 0.1.12
• matplotlib 1.3.1
• scipy 0.19.1

Reference

[1] Zhu, Jun-Yan, et al. "Unpaired image-to-image translation using cycle-consistent adversarial networks." arXiv preprint arXiv:1703.10593 (2017).

(Full paper: https://arxiv.org/pdf/1703.10593.pdf)