Text-to-Video Retrieval on MSR-VTT (Full)
34.3R@1VideoCoCa
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| VideoCoCaZero-shot=true2022.12 | 34.3 | 57.8 | 67 | — | — | |
| MDMMT-2 test meanSplit=full2022.03 | 33.7 | 60.5 | 70.8 | 3 | 37.8 | |
| MDMMT-2Split=full2022.03 | 33.4 | 60.1 | 70.5 | 3 | 39.2 | |
| MDMMT-22022.03 | 33.3 | 59.8 | 70.2 | 3 | 38.7 | |
| CLIP2TVSplit=full2022.03 | 33.1 | 58.9 | 68.9 | 3 | 44.7 | |
| CAMoESplit=full2022.03 | 32.9 | 58.3 | 68.4 | 3 | 42.6 | |
| VideoCoCaZero-shot=true, w.o. videos=true2022.12 | 32.9 | 56 | 65 | — | — | |
| CLIP2TVBackbone=ViT-B/162021.11 | 32.4 | 58.2 | 68.6 | 3 | 43.6 | |
| CoCaZero-shot=true2022.12 | 30 | 52.4 | 61.6 | — | — | |
| CLIP2TVSD=true2021.11 | 29.9 | 55.6 | 66.3 | 4 | 48.1 | |
| CLIP2Video2021.11 | 29.8 | 55.5 | 66.2 | 4 | 45.5 | |
| CLIP2VideoTest-set=Full2021.06 | 29.8 | 55.5 | 66.2 | 4 | 45.5 | |
| CLIP2VideoSplit=full2022.03 | 29.8 | 55.5 | 66.2 | 4 | 45.4 | |
| CLIP2TV2021.11 | 29.6 | 55.4 | 66 | 4 | 48.2 | |
| LAFFSplit=full2022.03 | 29.1 | 54.9 | 65.8 | — | — | |
| TACoSplit=full2022.03 | 24.8 | 52.1 | 64 | 5 | — | |
| CLIPZero-shot=true2022.12 | 23.3 | 44.2 | 53.6 | — | — | |
| MDMMT2021.11 | 23.1 | 49.8 | 61.8 | 6 | 52.8 | |
| MDMMTTest-set=Full2021.06 | 23.1 | 49.8 | 61.8 | 6 | 52.8 | |
| MDMMTSplit=full2022.03 | 23.1 | 49.8 | 61.8 | 6 | 52.8 | |
| MDMMT2022.03 | 22.8 | 49.5 | 61.5 | 6 | 53.8 | |
| CLIP aggSplit=full2022.03 | 21.5 | 41.1 | 50.4 | 4 | — | |
| CLIPTraining=WIT2021.02 | 21.4 | 41.1 | 50.4 | 10 | — | |
| CLIP-straight2021.11 | 21.4 | 41.1 | 50.4 | 10 | — | |
| CLIPTest-set=Full2021.06 | 21.4 | 41.1 | 50.4 | 10 | — | |
| UNiVLTest-set=Full2021.06 | 21.2 | 49.6 | 63.1 | 6 | — | |
| CLIPSplit=full2022.03 | 15.1 | 31.8 | 40.4 | 21 | 184.2 | |
| HT-pretrained2021.11 | 14.9 | 40.2 | 52.8 | 9 | — | |
| HT-pretrainedTest-set=Full2021.06 | 14.9 | 40.2 | 52.8 | 9 | — | |
| MMTSplit=full2022.03 | 10.7 | 31.1 | 43.4 | 15 | 88.2 | |
| MMT2022.03 | 10.4 | 30.2 | 42.3 | 16 | 89.4 | |
| CETest-set=Full2019.07 | 10 | 29 | 41.2 | 16 | 86.8 | |
| CETraining=MSR-VTT2021.02 | 10 | 29 | 42.2 | 16 | — | |
| CE2021.11 | 10 | 29 | 41.2 | 16 | 86.2 | |
| CETest-set=Full2021.06 | 10 | 29 | 41.2 | 16 | 86.8 | |
| CESplit=full2022.03 | 10 | 29 | 41.2 | 16 | 86.8 | |
| E2ETest-set=Full2019.07 | 9.9 | 24 | 32.4 | 29.5 | — | |
| E2ETraining=MSR-VTT2021.02 | 9.9 | 24 | 32.4 | 29.5 | — | |
| E2ETest-set=Full2021.06 | 9.9 | 24 | 32.4 | 29.5 | — | |
| Dual Enc.Test-set=Full2019.07 | 7.7 | 22 | 31.8 | 32 | — | |
| Dual Enc.Training=MSR-VTT2021.02 | 7.7 | 22 | 31.8 | 32 | — | |
| Dual Enc.Test-set=Full2021.06 | 7.7 | 22 | 31.8 | 32 | — | |
| Dual Enc.Split=full2022.03 | 7.7 | 22 | 31.8 | 32 | — | |
| Mithun et al.Test-set=Full2019.07 | 7 | 20.9 | 29.7 | 38 | 213.8 | |
| Multi CuesTraining=MSR-VTT2021.02 | 7 | 20.9 | 29.7 | 38 | — | |
| Multi CuesSplit=full2022.03 | 7 | 20.9 | 29.7 | 38 | — | |
| W2VVTest-set=Full2019.07 | 6.1 | 18.7 | 27.5 | 45 | — | |
| W2VVTraining=MSR-VTT2021.02 | 6.1 | 18.7 | 27.5 | 45 | — | |
| W2VVSplit=full2022.03 | 6.1 | 18.7 | 27.5 | 45 | — | |
| VSE++Test-set=Full2019.07 | 5.7 | 17.1 | 24.8 | 65 | 300.8 | |
| VSE++Training=MSR-VTT2021.02 | 5.7 | 17.1 | 24.8 | 65 | — | |
| VSE++Split=full2022.03 | 5.7 | 17.1 | 24.8 | 65 | — | |
| VSETest-set=Full2019.07 | 5 | 16.4 | 24.6 | 47 | 215.1 | |
| VSETraining=MSR-VTT2021.02 | 5 | 16.4 | 24.6 | 47 | — | |
| VSESplit=full2022.03 | 5 | 16.4 | 24.6 | 47 | — |