Feature Quantization Improves GAN Training

5 Apr 2020Yang ZhaoChunyuan LiPing YuJianfeng GaoChangyou Chen

The instability in GAN training has been a long-standing problem despite remarkable research efforts. We identify that instability issues stem from difficulties of performing feature matching with mini-batch statistics, due to a fragile balance between the fixed target distribution and the progressively generated distribution... (read more)

PDF Abstract

Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK BENCHMARK
Image-to-Image Translation anime-to-selfie FQ-GAN Kernel Inception Distance 10.23 # 1
Conditional Image Generation CIFAR-10 FQ-GAN Inception score 8.50 # 7
FID 5.34 # 2
Conditional Image Generation CIFAR-100 FQ-GAN Inception Score 9.74 # 2
FID 7.15 # 1
Image Generation FFHQ FQ-GAN FID 3.19 # 2
Conditional Image Generation ImageNet 128x128 FQ-GAN FID 13.77 # 6
Inception score 54.36 # 6
Conditional Image Generation ImageNet64x64 FQ-GAN Inception Score 25.96 # 1
FID 9.67 # 1
Image-to-Image Translation selfie-to-anime FQ-GAN Kernel Inception Distance 11.40 # 1

Methods used in the Paper