Video Moment Retrieval on ActivityNet-Captions
51.68R1@0.5 RecallG2L
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| G2LText Feat=BERT, Video Feat=C3D2023.06 | 51.68 | — | 33.35 | — | |
| GTRText Feat=Glove, Video Feat=C3D2023.06 | 50.57 | — | 29.11 | — | |
| BM-DETRText Feat=BERT, Video Feat=C3D2023.06 | 49.98 | — | 30.88 | — | |
| BM-DETRText Feat=Glove, Video Feat=C3D2023.06 | 49.62 | — | 30.61 | — | |
| MMNText Feat=DistilBERT, Video Feat=C3D2023.06 | 48.59 | — | 29.26 | — | |
| MMNVideo Features=C3D, Text Features=Glove, Training time=1.5h, Inference time=662.12s2023.07 | 48.26 | 65.21 | 28.95 | 28.74 | |
| CBLNText Feat=Glove, Video Feat=C3D2023.06 | 48.12 | — | 27.6 | — | |
| SSCSText Feat=Glove, Video Feat=C3D2023.06 | 46.67 | — | 27.56 | — | |
| MomentDiffVideo Features=C3D, Text Features=Glove, Training time=0.06h, Inference time=91.43s2023.07 | 46.52 | 62.79 | 28.43 | 28.19 | |
| DRNText Feat=Glove, Video Feat=C3D2023.06 | 45.45 | — | 24.39 | — | |
| 2DTANVideo Features=C3D, Text Features=Glove, Training time=1.1h, Inference time=523.74s2023.07 | 44.63 | 59.92 | 27.53 | 27.26 | |
| 2D-TANText Feat=Glove, Video Feat=C3D2023.06 | 44.51 | — | 26.54 | — | |
| IVG-DCLText Feat=Glove, Video Feat=C3D2023.06 | 43.84 | — | 27.1 | — | |
| DeNetText Feat=Glove, Video Feat=C3D2023.06 | 43.79 | — | — | — | |
| VSLNetText Feat=Glove, Video Feat=C3D2023.06 | 43.22 | — | 26.16 | — | |
| MomentDETRVideo Features=C3D, Text Features=Glove, Training time=0.05h, Inference time=52.72s2023.07 | 43.19 | 61.87 | 25.74 | 25.63 |