Image Segmentation on SA-1B (test)
60.8ARSAM (supervised)
Evaluation results
| Method | Links | |
|---|---|---|
| SAM (supervised)Backbone=ViT-B/8, Number of Parameters=85M, Number of training images=11M, Zero-shot=true2024.06 | 60.8 | |
| UnSAMBackbone=RN-50, Number of Parameters=23M, Number of training images=0.4M, Zero-shot=true2024.06 | 44.5 | |
| UnSAMBackbone=RN-50, Number of Parameters=23M, Number of training images=0.2M, Zero-shot=true2024.06 | 43.6 | |
| UnSAMBackbone=RN-50, Number of Parameters=23M, Number of training images=0.1M, Zero-shot=true2024.06 | 41.9 | |
| SOHESBackbone=ViT-B/8, Number of Parameters=85M, Number of training images=0.2M, Zero-shot=true2024.06 | 33.3 | |
| CutLERBackbone=RN-50, Number of Parameters=23M, Number of training images=1.3M, Zero-shot=true2024.06 | 17 | |
| FreeSOLOBackbone=RN-101, Number of Parameters=45M, Number of training images=1.3M, Zero-shot=true2024.06 | 2.2 |