PHStarGANv2 n=4

Reported on 6 benchmarks across 3 tasks · 1 paper · 3 SOTA

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

Computer Vision2 results

Image-to-Image TranslationonCelebA-HQ
LPIPS· 2022-05-04
0.29
SOTA
Hypercomplex Image-to-Image Translation arXiv:2205.02087
Image-to-Image TranslationonCelebA-HQ
FID· 2022-05-04
16.54
best: 13.73 (StarGAN v2)
Hypercomplex Image-to-Image Translation arXiv:2205.02087

Medical2 results

Image GenerationonCelebA-HQ
LPIPS· 2022-05-04
0.29
SOTA
Hypercomplex Image-to-Image Translation arXiv:2205.02087
Image GenerationonCelebA-HQ
FID· 2022-05-04
16.54
best: 4.86 (MAT)
Hypercomplex Image-to-Image Translation arXiv:2205.02087

Miscellaneous2 results

1 Image, 2*2 StitchingonCelebA-HQ
LPIPS· 2022-05-04
0.29
SOTA
Hypercomplex Image-to-Image Translation arXiv:2205.02087
1 Image, 2*2 StitchingonCelebA-HQ
FID· 2022-05-04
16.54
best: 13.73 (StarGAN v2)
Hypercomplex Image-to-Image Translation arXiv:2205.02087