Multimodal Image-to-Image Translation (RGB+Shade to Normal) on Taskonomy
0.58MAETokenFusion (S)
Evaluation results
| Method | Links | ||
|---|---|---|---|
| TokenFusion (S)Architecture=Transformer-based, Scale=Small, Resolution=256x2562022.04 | 0.58 | 1.51 | |
| TokenFusion (Ti)Architecture=Transformer-based, Scale=Tiny, Resolution=256x2562022.04 | 0.73 | 1.83 | |
| CENArchitecture=CNN-based, Resolution=256x2562022.04 | 1.12 | 2.51 | |
| Concat (S)Architecture=Transformer-based, Scale=Small, Resolution=256x2562022.04 | 1.18 | 2.73 | |
| Self-Att.Architecture=CNN-based, Resolution=256x2562022.04 | 1.26 | 2.76 | |
| ConcatArchitecture=CNN-based, Resolution=256x2562022.04 | 1.34 | 2.85 | |
| Align.Architecture=CNN-based, Resolution=256x2562022.04 | 1.52 | 3.25 | |
| Concat (Ti)Architecture=Transformer-based, Scale=Tiny, Resolution=256x2562022.04 | 1.52 | 3.15 |