Video-to-Text Retrieval on Synthetic Time-order Consistency 1.0 (test)
52Recall@1VindLU
Evaluation Results
| Method | Links | |
|---|---|---|
| VindLUParadigm=Video-Language Contrastive2023.01 | 52 | |
| CLIP4ClipParadigm=Image-Language adapted to video2023.01 | 51.1 | |
| VideoCLIPParadigm=Video-Language Contrastive2023.01 | 51.1 | |
| BridgeFormerParadigm=Video-Language Masking2023.01 | 51.1 | |
| ChanceParadigm=-2023.01 | 50 | |
| Frozen in TimeParadigm=Video-Language Contrastive2023.01 | 49.4 | |
| CLIP2VideoParadigm=Image-Language adapted to video2023.01 | 47.8 | |
| CenterCLIPParadigm=Image-Language adapted to video2023.01 | 46.1 |