Reproduce Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks
-
Adversarial Loss
-
Cycle consistency Loss
-
Identity Loss
- 이게 없어지니까 학습속도가 더뎌짐
- Paint에서 color composition을 유지하기 위해 사용
- Input image의 color보존에 도움을 줌
Reproduce CycleGAN using pytorch
Reproduce Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks
Adversarial Loss
Cycle consistency Loss
Identity Loss
Reproduce CycleGAN using pytorch