Video Paragraph Captioning on ActivityNet Captions ae (val)
18.16METEORVLTinT
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| VLTinTInput=C3D/Ling2022.11 | 18.16 | 14.93 | 33.07 | 4.87 | 36.86 | 77.72 | — | |
| VLCapInput=C3D + Language2022.06 | 17.78 | 14 | 32.58 | 4.42 | 36.37 | 78.01 | — | |
| Vid2SeqBackbone=V (CLIP), Proposals=Learnt2023.02 | 17 | — | 28 | — | — | — | — | |
| AdvInfuse_detection_features=true, sentence_level_recurrence=true2020.05 | 16.6 | 10.04 | 20.97 | 5.76 | — | — | — | |
| AdvInfYear=CVPR2019, Input=C3D + Object2022.06 | 16.6 | 10.04 | 20.97 | 5.76 | — | — | — | |
| AdvInfVenue=CVPR, Input=C3D/Object2022.11 | 16.6 | 10.04 | 20.97 | 5.76 | — | — | — | |
| AdvInfBackbone=V (ResNet-101) + F + O, Proposals=Ground Truth2023.02 | 16.6 | — | 21 | — | — | — | — | |
| TextKGEvaluation Mode=paragraph-level2023.03 | 16.5 | 11.3 | 26.6 | — | — | — | 6.3 | |
| GVDsupuse_detection_features=true, sentence_level_recurrence=false2020.05 | 16.41 | 11.3 | 22.94 | 7.04 | — | — | — | |
| GVDSupBackbone=V (ResNet-101) + F + O, Proposals=Ground Truth2023.02 | 16.4 | — | 22.9 | — | — | — | — | |
| GVDsupEvaluation Mode=paragraph-level2023.03 | 16.4 | 11.3 | 22.9 | — | — | — | 7 | |
| PDVCYear=ICCV2021, Input=C3D + Flow2022.06 | 15.93 | 11.8 | 27.27 | — | — | — | — | |
| PDVCVenue=ICCV, Input=C3D/Flow2022.11 | 15.93 | 11.8 | 27.27 | — | — | — | — | |
| PDVCBackbone=V + F (TSN), Proposals=Ground Truth2023.02 | 15.9 | — | 27.3 | — | — | — | — | |
| PDVCBackbone=V (CLIP), Proposals=Learnt, Re-evaluated using official codebase=true2023.02 | 15.9 | — | 23.6 | — | — | — | — | |
| PDVCBackbone=V + F (TSN), Proposals=Learnt2023.02 | 15.8 | — | 20.5 | — | — | — | — | |
| GVDuse_detection_features=true, sentence_level_recurrence=false2020.05 | 15.71 | 11.04 | 21.95 | 8.76 | — | — | — | |
| GVDYear=CVPR2019, Input=Res200 + Flow + Object2022.06 | 15.71 | 11.04 | 21.95 | 8.76 | — | — | — | |
| GVDVenue=CVPR, Input=Res200/Flow/Object2022.11 | 15.71 | 11.04 | 21.95 | 8.76 | — | — | — | |
| MARTBackbone=V (ResNet-200) + F, Proposals=Ground Truth2023.02 | 15.7 | — | 23.4 | — | — | — | — | |
| GVDEvaluation Mode=paragraph-level2023.03 | 15.7 | 11 | 22 | — | — | — | 8.8 | |
| MARTEvaluation Mode=paragraph-level2023.03 | 15.7 | 10.3 | 23.4 | — | — | — | 5.2 | |
| MARTuse_detection_features=false, sentence_level_recurrence=true2020.05 | 15.68 | 10.33 | 23.42 | 5.18 | — | — | — | |
| MARTYear=ACL2020, Input=Res200 + Flow2022.06 | 15.68 | 10.33 | 23.42 | 5.18 | 30.32 | 75.71 | — | |
| MARTVenue=ACL, Input=Res200/Flow2022.11 | 15.68 | 10.33 | 23.42 | 5.18 | 30.32 | 75.71 | — | |
| VTransformeruse_detection_features=false, sentence_level_recurrence=false2020.05 | 15.64 | 9.75 | 22.16 | 7.79 | — | — | — | |
| Vanilla TransformerYear=CVPR2018, Input=Res200 + Flow2022.06 | 15.64 | 9.75 | 22.16 | 7.79 | 28.9 | 77.4 | — | |
| Vanilla Trans.Venue=CVPR, Input=Res200/Flow2022.11 | 15.64 | 9.75 | 22.16 | 7.79 | 28.9 | 77.4 | — | |
| VTransformerBackbone=V (ResNet-200) + F, Proposals=Ground Truth2023.02 | 15.6 | — | 22.2 | — | — | — | — | |
| Van-TransEvaluation Mode=paragraph-level2023.03 | 15.6 | 9.8 | 22.2 | — | — | — | 7.8 | |
| Transformer-XLBackbone=V (ResNet-200) + F, Proposals=Ground Truth2023.02 | 15.1 | — | 21.7 | — | — | — | — | |
| Trans-XLEvaluation Mode=paragraph-level2023.03 | 15.1 | 10.4 | 21.7 | — | — | — | 8.5 | |
| Transformer-XLuse_detection_features=false, sentence_level_recurrence=true2020.05 | 15.09 | 10.39 | 21.67 | 8.54 | — | — | — | |
| Transformer-XLYear=ACL2019, Input=Res200 + Flow2022.06 | 15.09 | 10.39 | 21.67 | 8.54 | 30.18 | 75.96 | — | |
| Trans.-XLVenue=ACL, Input=Res200/Flow2022.11 | 15.09 | 10.39 | 21.67 | 8.54 | 30.18 | 75.96 | — | |
| Trans-XLRGEvaluation Mode=paragraph-level2023.03 | 14.8 | 10.2 | 20.4 | — | — | — | 8.9 | |
| Transformer-XLRGuse_detection_features=false, sentence_level_recurrence=true2020.05 | 14.77 | 10.17 | 20.4 | 8.85 | — | — | — | |
| Transformer-XLRGYear=ACL2020, Input=Res200 + Flow2022.06 | 14.77 | 10.17 | 20.4 | 8.85 | — | — | — | |
| Trans.-XLRGVenue=ACL, Input=Res200/Flow2022.11 | 14.77 | 10.17 | 20.4 | 8.85 | — | — | — | |
| MFTuse_detection_features=false, sentence_level_recurrence=true2020.05 | 14.73 | 10.29 | 19.12 | 17.71 | — | — | — | |
| MFTBackbone=V + F (TSN), Proposals=Learnt2023.02 | 14.7 | — | 19.1 | — | — | — | — | |
| HSEEvaluation Mode=paragraph-level2023.03 | 13.8 | 9.8 | 18.8 | — | — | — | 13.2 | |
| HSEuse_detection_features=false, sentence_level_recurrence=true2020.05 | 13.78 | 9.84 | 18.78 | 13.22 | — | — | — | |
| Human2020.05 | — | — | — | 0.98 | — | — | — |