Temporal clip retrieval (text-to-video) on YouCook
47.2R@5TrajViT
Evaluation Results
| Method | Links | |
|---|---|---|
| TrajViTTraining subset=4M, Inference frames=162025.05 | 47.2 | |
| ViT3DTraining subset=4M, Inference frames=162025.05 | 46.8 | |
| RLTTraining subset=4M, Inference frames=162025.05 | 46.2 | |
| TokenLearnerTraining subset=4M, Inference frames=162025.05 | 46 | |
| ViViTTraining subset=4M, Inference frames=162025.05 | 45.7 | |
| AutoMergeTraining subset=4M, Inference frames=162025.05 | 45.7 | |
| ToMeTraining subset=4M, Inference frames=162025.05 | 45.4 |