StyleGAN2 + DiffAugment + D2D-CE

Reported on 4 benchmarks across 2 tasks · 1 paper · 4 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Medical2 results

Image GenerationonCIFAR-10
FID· 2021-11-01
2.26
best: 1.22 (GMem)
SOTA
Rebooting ACGAN: Auxiliary Classifier GANs with Stable Training arXiv:2111.01118
Image GenerationonCIFAR-10
Inception score· 2021-11-01
10.51
SOTA
Rebooting ACGAN: Auxiliary Classifier GANs with Stable Training arXiv:2111.01118

Computer Vision2 results

Conditional Image GenerationonCIFAR-10
FID· 2021-11-01
2.26
best: 1.64 (EDM-G++ (conditional))
SOTA
Rebooting ACGAN: Auxiliary Classifier GANs with Stable Training arXiv:2111.01118
Conditional Image GenerationonCIFAR-10
Inception score· 2021-11-01
10.51
SOTA
Rebooting ACGAN: Auxiliary Classifier GANs with Stable Training arXiv:2111.01118