Linguistic Promptable Video Object Segmentation on Uni-EndoVis I 18
85.1J&FUniSurgSAM
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| UniSurgSAMSetting=Online, Modality=Textual2026.04 | 85.1 | 85.3 | 84.9 | 55 | |
| UniSurgSAMSetting=Online, Modality=Audio2026.04 | 83.2 | 83.4 | 82.9 | 53 | |
| ResurgSAM2Setting=Online, Modality=Textual2026.04 | 81.1 | 81.5 | 80.7 | 58 | |
| RefSAMSetting=Online, Modality=Textual2026.04 | 72.9 | 73.4 | 72.3 | 25 | |
| OnlineReferSetting=Online, Modality=Textual2026.04 | 72.2 | 71.9 | 72.5 | 26 | |
| ReferFormerSetting=Offline, Modality=Textual2026.04 | 71.1 | 71 | 71.2 | 42 | |
| SAMWISESetting=Semi-online, Modality=Textual2026.04 | 70.6 | 69.8 | 71.4 | 12 | |
| SurgRefSetting=Offline, Modality=Textual2026.04 | 70.3 | 70.4 | 70.2 | 40 | |
| RSVISSetting=Semi-online, Modality=Textual2026.04 | 68.4 | 68.6 | 68.2 | 28 | |
| MUTRSetting=Offline, Modality=Textual2026.04 | 67.6 | 67.8 | 67.3 | 32 |