Text-to-Video Retrieval on MSR-VTT 9K (train)
44.5R@1CLIP4Clip-seqTransf
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| CLIP4Clip-seqTransfTrainD=W+M, E2E=false2021.04 | 44.5 | 71.4 | 81.6 | 2 | 15.3 | |
| CLIP4Clip-meanPTrainD=W+M, E2E=false2021.04 | 43.1 | 70.4 | 80.8 | 2 | 16.2 | |
| CLIP4Clip-seqLSTMTrainD=W+M, E2E=false2021.04 | 42.5 | 70.8 | 80.7 | 2 | 16.7 | |
| CLIP4Clip-tightTransfTrainD=W+M, E2E=true2021.04 | 40.2 | 71.5 | 80.5 | 2 | 13.4 | |
| MDMMTTrainD=MD+M, E2E=false2021.04 | 38.9 | 69 | 79.7 | 2 | 16.5 | |
| FrozenTrainD=CW+M, E2E=true2021.04 | 31 | 59.5 | 70.5 | 3 | — | |
| HiTTrainD=H+M, E2E=false2021.04 | 30.7 | 60.9 | 73.2 | 2.6 | — | |
| SSBTrainD=H+M, E2E=false2021.04 | 30.1 | 58.5 | 69.3 | 3 | — | |
| TT-CE+TrainD=M, E2E=false2021.04 | 29.6 | 61.6 | 74.2 | 3 | — | |
| AVLnetTrainD=H+M, E2E=false2021.04 | 27.1 | 55.6 | 66.6 | 4 | — | |
| MMTPTrainD=H+M, E2E=false2021.04 | 26.6 | 57.1 | 69.6 | 4 | 24 | |
| CETrainD=M, E2E=false2021.04 | 20.9 | 48.8 | 62.4 | 6 | 28.2 |