LSGM

Reported on 3 benchmarks across 1 task · 1 paper · 2 SOTA

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

Medical3 results

Image GenerationonCelebA 256x256
FID· 2021-06-10
7.22
best: 3.25 (StyleSwin)
SOTA
Score-based Generative Modeling in Latent Space arXiv:2106.05931
Image GenerationonCelebA-HQ 256x256
FID· 2021-06-10
7.22
best: 3.15 (RDM)
SOTA
Score-based Generative Modeling in Latent Space arXiv:2106.05931
Image GenerationonCelebA 256x256
bpd· 2021-06-10
0.7
best: 0.51 (Efficient-VDVAE)
Score-based Generative Modeling in Latent Space arXiv:2106.05931