Semantic Segmentation on YCB-V
91.8mIoUFull FT (SAM 2-L)
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Full FT (SAM 2-L)Params=224M2026.05 | 91.8 | — | — | |
| Full FT (SAM ViT-H)Params=632M2026.05 | 90.5 | — | — | |
| RepSAMParams=4.0M2026.05 | 88 | — | — | |
| HQ-SAMParams=610M2026.05 | 81.7 | — | — | |
| VeRAParams=0.9M2026.05 | 81 | — | — | |
| DoRAParams=4.1M2026.05 | 80.8 | — | — | |
| AdaLoRAParams=4.5M2026.05 | 78.9 | — | — | |
| LoRA (uniform r=8)Params=4.2M2026.05 | 73.9 | — | — | |
| FastSAMParams=68M2026.05 | 73 | — | — | |
| SAM-DSAParams=5.1M2026.05 | 70.3 | — | — | |
| SAM-AdapterParams=4.8M2026.05 | 68.1 | — | — | |
| EfficientSAMParams=13M2026.05 | 62.9 | — | — | |
| MobileSAMParams=10M2026.05 | 60.1 | — | — | |
| SAM 2-L (zero-shot)Params=224M2026.05 | 48.9 | — | — | |
| SAM ViT-H (zero-shot)Params=—2026.05 | 42.5 | — | — | |
| OCH3ROpen-vocabulary=true2026.05 | 9.14 | 72 | 80.03 | |
| MAFT+Open-vocabulary=true2026.05 | 5 | 12.16 | 79.01 | |
| ODISEOpen-vocabulary=true2026.05 | 4.51 | 12.17 | 76.76 | |
| SANOpen-vocabulary=true2026.05 | 4.25 | 44.77 | 72.83 | |
| FC-CLIPOpen-vocabulary=true2026.05 | 3.69 | 12.15 | 69.96 | |
| LSegOpen-vocabulary=true2026.05 | 2.63 | 12.41 | 23.05 | |
| CAT-SegOpen-vocabulary=true2026.05 | 2.38 | 12.12 | 65.78 | |
| OVSegOpen-vocabulary=true2026.05 | 1.29 | 12.2 | 58.66 |