Text-to-Video Retrieval on SSv2-Events (zero-shot)
11.4mAPOwl-Con
Evaluation results
| Method | Links | |
|---|---|---|
| Owl-Conzero-shot=true2023.11 | 11.4 | |
| End-to-End VNLIzero-shot=true2023.11 | 10.4 | |
| Owl-Randzero-shot=true2023.11 | 9.9 | |
| TACTzero-shot=true2023.11 | 7.8 | |
| Owl-Basezero-shot=true2023.11 | 6.8 | |
| VideoCLIPzero-shot=true2023.11 | 6.4 | |
| ImageBind (video language)zero-shot=true2023.11 | 5.5 | |
| Randomzero-shot=true2023.11 | 3.3 |