Semi-supervised Video Object Segmentation on MOSE (val)
77.9J&F ScoreSAM 2
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| SAM 2Backbone=Hiera-L, Extra Training Data (SA-1B, SA-V)=false2024.08 | 77.9 | 73.9 | 81.9 | |
| SAM 2Backbone=Hiera-B+, Extra Training Data (SA-1B, SA-V)=false2024.08 | 76.6 | 72.6 | 80.6 | |
| SAM 2Backbone=Hiera-L, Extra Training Data (SA-1B, SA-V)=true2024.08 | 74.6 | 70.6 | 78.6 | |
| SAM 2Backbone=Hiera-B+, Extra Training Data (SA-1B, SA-V)=true2024.08 | 73.8 | 69.7 | 77.9 | |
| SAM 2Backbone=Hiera-S, Extra Training Data (SA-1B, SA-V)=true2024.08 | 73.5 | 69.2 | 77.7 | |
| SAM 2Backbone=Hiera-T, Extra Training Data (SA-1B, SA-V)=true2024.08 | 71.8 | 67.4 | 76.1 | |
| Cutie-base+2024.08 | 71.7 | 67.6 | 75.8 | |
| Cutie-base2024.08 | 69.9 | 65.8 | 74.1 | |
| DEVA2024.08 | 66 | 61.8 | 70.3 | |
| SwinB-DeAOT-L2024.08 | 59.9 | 55.7 | 64 | |
| XMem2024.08 | 59.6 | 55.4 | 63.7 | |
| SwinB-AOT-L2024.08 | 59.4 | 55.5 | 63.2 | |
| STCN2024.08 | 52.5 | 48.5 | 56.6 | |
| RDE2024.08 | 46.8 | 42.4 | 51.3 |