Classification on VB100
20.7AccuracyTri-CLIP
Evaluation Results
| Method | Links | |
|---|---|---|
| Tri-CLIPInput Modality=Vision, Evaluation Setting=Zero-shot, Evaluation Protocol=Multi-frame2025.11 | 20.7 | |
| CLIPInput Modality=Vision, Evaluation Setting=Zero-shot, Evaluation Protocol=Multi-frame2025.11 | 20.6 | |
| ConFuInput Modality=Vision, Evaluation Setting=Zero-shot, Evaluation Protocol=Multi-frame2025.11 | 19.3 | |
| ConFuInput Modality=Audio-Visual (A+V), Evaluation Setting=Zero-shot, Evaluation Protocol=Multi-frame2025.11 | 18.1 | |
| GRAMInput Modality=Vision, Evaluation Setting=Zero-shot, Evaluation Protocol=Multi-frame2025.11 | 13.7 | |
| SymileInput Modality=Audio-Visual (A+V), Evaluation Setting=Zero-shot, Evaluation Protocol=Multi-frame2025.11 | 13.4 | |
| TRIANGLEInput Modality=Audio-Visual (A+V), Evaluation Setting=Zero-shot, Evaluation Protocol=Multi-frame2025.11 | 12.1 | |
| GRAMInput Modality=Audio-Visual (A+V), Evaluation Setting=Zero-shot, Evaluation Protocol=Multi-frame2025.11 | 8 | |
| CLIPInput Modality=Audio, Evaluation Setting=Zero-shot, Evaluation Protocol=Multi-frame2025.11 | 4.2 | |
| Tri-CLIPInput Modality=Audio, Evaluation Setting=Zero-shot, Evaluation Protocol=Multi-frame2025.11 | 3.9 | |
| ConFuInput Modality=Audio, Evaluation Setting=Zero-shot, Evaluation Protocol=Multi-frame2025.11 | 3.4 | |
| GRAMInput Modality=Audio, Evaluation Setting=Zero-shot, Evaluation Protocol=Multi-frame2025.11 | 1.3 |