Dialogue Scene Segmentation on VSTAR (test)
53.6mIoUSWST
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| SWSTmodality=multimodal (text + video), sampled frames=32023.05 | 53.6 | 54.3 | 50.3 | |
| SWSTmodality=video-only, sampled frames=32023.05 | 48.1 | 47.4 | 43 | |
| Bassal2023.05 | 46.6 | 44.2 | 40.1 | |
| SWSTmodality=text-only2023.05 | 45.3 | 35.1 | 38 | |
| SWSTmodality=video-only, sampled frames=12023.05 | 44.8 | 41.9 | 38.5 | |
| ShotCol2023.05 | 42.7 | 41.2 | 36.5 | |
| Random2023.05 | 25.1 | 6 | 7.5 |