Human Parts Segmentation on PASCAL Human Parts OOD Corruptions (test)
61.73Fog AccFTP
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| FTPBackbone=Swin-Tiny, Pre-trained=ImageNet21K, Time (s/it)=0.3972023.10 | 61.73 | 44.97 | 40.55 | 61.23 | 52.12 | 11.86 | |
| TPGMBackbone=Swin-Tiny, Pre-trained=ImageNet21K, Time (s/it)=0.6022023.10 | 60.16 | 46.91 | 37.3 | 59.81 | 51.04 | 9.55 | |
| LLRDBackbone=Swin-Tiny, Pre-trained=ImageNet21K, Time (s/it)=0.2782023.10 | 60.1 | 44.61 | 36.9 | 59.84 | 50.36 | 8.09 | |
| L2-SPBackbone=Swin-Tiny, Pre-trained=ImageNet21K, Time (s/it)=0.3362023.10 | 58.46 | 45.35 | 34.36 | 58.4 | 49.14 | 5.47 | |
| MARS-SPBackbone=Swin-Tiny, Pre-trained=ImageNet21K, Time (s/it)=0.3082023.10 | 58.04 | 42.51 | 32.66 | 58.33 | 47.89 | 2.77 | |
| Vanilla FTBackbone=Swin-Tiny, Pre-trained=ImageNet21K, Time (s/it)=0.282023.10 | 57.5 | 40.76 | 30.64 | 57.47 | 46.59 | 0 | |
| AdapterBackbone=Swin-Tiny, Pre-trained=ImageNet21K, Time (s/it)=0.2212023.10 | 57.11 | 45.03 | 33.12 | 57.25 | 48.13 | 3.3 | |
| BitFitBackbone=Swin-Tiny, Pre-trained=ImageNet21K, Time (s/it)=0.2352023.10 | 55.66 | 45.81 | 32.18 | 55.89 | 47.39 | 1.7 |