HiFill (20-30% free form)

Reported on 4 benchmarks across 2 tasks · 1 paper

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

Medical2 results

Image GenerationonPlaces2 val
FID· 2020-05-19
15.7
best: 11.7 (Palatte (20-30% free form))
Contextual Residual Aggregation for Ultra High-Resolution Image Inpainting arXiv:2005.09704
Image GenerationonPlaces2 val
PD· 2020-05-19
92.8
best: 115.4 (HiFill (128×128 center mask))
Contextual Residual Aggregation for Ultra High-Resolution Image Inpainting arXiv:2005.09704

Computer Vision2 results

Image InpaintingonPlaces2 val
FID· 2020-05-19
15.7
best: 11.7 (Palatte (20-30% free form))
Contextual Residual Aggregation for Ultra High-Resolution Image Inpainting arXiv:2005.09704
Image InpaintingonPlaces2 val
PD· 2020-05-19
92.8
best: 115.4 (HiFill (128×128 center mask))
Contextual Residual Aggregation for Ultra High-Resolution Image Inpainting arXiv:2005.09704