Video-to-Text Retrieval on LSMDC Movie Description (test)
24.5R@1CenterCLIP (k-medoids++, Bt = 4, 160)
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| CenterCLIP (k-medoids++, Bt = 4, 160)Backbone=ViT-B/16, MeM. GB=17.6, speed ms=86.52022.05 | 24.5 | 46.4 | 55.8 | 7 | 41.3 | |
| CLIP4clip (meanP)Backbone=ViT-B/16, MeM. GB=25.7, speed ms=59.62022.05 | 22.5 | 42.9 | 53.5 | 9 | 45.1 | |
| CenterCLIP (k-medoids++, Bt = 4, 49)Backbone=ViT-B/32, MeM. GB=15, speed ms=22.92022.05 | 21.4 | 40.3 | 50.8 | 10 | 48.4 | |
| CLIP4clip (meanP)Backbone=ViT-B/32, MeM. GB=20.8, speed ms=24.42022.05 | 21.2 | 39.3 | 48.4 | 12 | 50.8 | |
| CenterCLIP (k-medoids++, Bt = 6, 49)Backbone=ViT-B/32, MeM. GB=16.4, speed ms=23.92022.05 | 21.1 | 41.2 | 50.2 | 10 | 48.7 | |
| CLIP4clip (seqTransf)2022.05 | 20.8 | 39 | 48.6 | 12 | 54.2 | |
| CLIP4clip (meanP)MeM. GB=20.8, speed ms=24.42022.05 | 20.6 | 39.4 | 47.5 | 13 | 56.7 | |
| CenterCLIP (spectral, Bt = 6, 49)Backbone=ViT-B/32, MeM. GB=16.4, speed ms=40.82022.05 | 20.6 | 39.5 | 48.8 | 12 | 51.4 | |
| CenterCLIP (spectral, Bt = 4, 49)Backbone=ViT-B/32, MeM. GB=15, speed ms=43.62022.05 | 19.5 | 39.9 | 48 | 12 | 50.1 | |
| JSFusion2022.05 | 12.3 | 28.6 | 38.9 | 20 | — | |
| MMT2022.05 | 12.3 | 28.6 | 38.9 | 20 | 76 | |
| CLIP zero-shot2022.05 | 7.5 | 18.4 | 25.1 | 58 | 151 |