Metric: mAP (higher is better)
| # | Model↕ | mAP▼ | Extra Data | Paper | Date↕ | Code |
|---|---|---|---|---|---|---|
| 1 | LaViLa (Finetuned, TimeSformer-L) | 36.1 | No | Learning Video Representations from Large Langua... | 2022-12-08 | Code |
| 2 | EgoVLPv2 | 34.1 | No | EgoVLPv2: Egocentric Video-Language Pre-training... | 2023-07-11 | Code |
| 3 | HierVL | 33.8 | No | HierVL: Learning Hierarchical Video-Language Emb... | 2023-01-05 | Code |
| 4 | EgoVLP | 32.1 | No | Egocentric Video-Language Pretraining | 2022-06-03 | Code |
| 5 | LaViLa (Zero-shot, TimeSformer-L) | 28.9 | No | Learning Video Representations from Large Langua... | 2022-12-08 | Code |
| 6 | HierVL (Zero-shot) | 26 | No | HierVL: Learning Hierarchical Video-Language Emb... | 2023-01-05 | Code |