Open-vocabulary Video Instance Segmentation on OVIS (val)
60.5mAPSAM 3
Evaluation Results
| Method | Links | |
|---|---|---|
| SAM 32025.11 | 60.5 | |
| SAM 3 Detector + T-by-D2025.11 | 55.1 | |
| GLEEProcessing mode=all NPs at once2025.11 | 38.7 | |
| GLEEProcessing mode=one NP at a time2025.11 | 32.4 | |
| LLMDet + SAM 3 Tracker2025.11 | 20.4 |