Dense Video Captioning on YouCook2 (val)
80.53CIDErSTaRC
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| STaRCPT=✓2026.03 | 80.53 | 10.73 | — | — | — | — | 34.34 | |
| STaRCPretrained=true2026.03 | 80.53 | 10.73 | 13.86 | 6.75 | — | — | — | |
| Sali4VidPretrained=true2026.03 | 75.8 | 10.28 | 13.54 | 6.35 | — | — | — | |
| HiCM2Pretrained=true2026.03 | 71.84 | 10.73 | 12.8 | 6.11 | — | — | — | |
| HiCM²Pretrain=✓2026.04 | 71.8 | 10.7 | 12.8 | — | — | — | — | |
| Vid2SeqPretrained=true, reproduced_from_official=true2026.03 | 66.29 | 9.87 | 12.41 | 5.64 | — | — | — | |
| DIBSTraining paradigm=Fully-Supervised, Pre-trained=56k, Backbone=UniVL2024.04 | 59.35 | 7.97 | 9.41 | — | — | — | — | |
| Vid2SeqBackbone=CLIP2023.02 | 47.1 | 7.9 | 9.3 | — | — | — | — | |
| Vid2SeqTraining paradigm=Fully-Supervised, Pre-trained=15M, Backbone=CLIP2024.04 | 47.1 | 7.9 | 9.3 | — | — | — | — | |
| Vid2SeqPretrain=✓2026.04 | 47.1 | 7.9 | 9.3 | — | — | — | — | |
| PDVCTraining paradigm=Fully-Supervised, Pre-trained=None, Backbone=UniVL2024.04 | 46.02 | 6.87 | 7.87 | — | — | — | — | |
| DIBSTraining paradigm=Fully-Supervised, Pre-trained=56k, Backbone=CLIP2024.04 | 44.44 | 6.39 | 7.51 | — | — | — | — | |
| DIBSPretrained=true2026.03 | 44.44 | 6.39 | 7.51 | — | — | — | — | |
| TimeExpertPT=✓2026.03 | 39 | 7.2 | — | — | — | — | 33.5 | |
| TRACEPT=✓2026.03 | 35.5 | 6.7 | — | — | — | — | 31.8 | |
| CACMIPretrained=false2026.03 | 34.83 | 5.57 | 6.21 | 1.7 | — | — | — | |
| E2DVCPretrained=false2026.03 | 34.26 | 5.39 | 6.11 | 1.68 | — | — | — | |
| DenseStep2M pipelinePretrain=×2026.04 | 33.6 | 5.6 | 9.4 | — | — | — | — | |
| Streaming V2SPretrained=true2026.03 | 32.9 | 6 | 7.1 | — | — | — | — | |
| CM²Pretrain=×2026.04 | 31.7 | 5.3 | 6.1 | — | — | — | — | |
| CM2Pretrained=false2026.03 | 31.66 | 5.34 | 6.08 | 1.63 | — | — | — | |
| PDVCPretrained=false2026.03 | 29.69 | 4.92 | 5.56 | 1.4 | — | — | — | |
| DIBSTraining paradigm=Weakly-Supervised, Pre-trained=None, Backbone=UniVL2024.04 | 29.62 | 4.96 | 5.9 | — | — | — | — | |
| PDVCBackbone=CLIP2023.02 | 28.9 | 4.9 | 5.7 | — | — | — | — | |
| PDVCPretrain=×2026.04 | 28.9 | 4.9 | 5.7 | — | — | — | — | |
| PDVCTraining paradigm=Fully-Supervised, Pre-trained=None, Backbone=CLIP2024.04 | 28.37 | 5 | 5.47 | — | — | — | — | |
| GVLFeatures=TSN2023.03 | 26.52 | 4.91 | 5.01 | 1.04 | — | — | — | |
| E2ESGTraining paradigm=Fully-Supervised, Pre-trained=None, Backbone=C3D2024.04 | 25 | — | 3.49 | — | — | — | — | |
| E2ESGPretrain=×2026.04 | 25 | — | 3.5 | — | — | — | — | |
| PDVCTraining paradigm=Fully-Supervised, Pre-trained=None, Backbone=TSN2024.04 | 22.71 | 4.42 | 4.74 | — | — | — | — | |
| PDVCFeatures=TSN2023.03 | 22.71 | 4.42 | 4.74 | 0.8 | — | — | — | |
| PDVCBackbone=TSN2023.02 | 22.7 | 4.4 | 4.7 | — | — | — | — | |
| DIBSTraining paradigm=Weakly-Supervised, Pre-trained=None, Backbone=CLIP2024.04 | 22.05 | 4.03 | 4.63 | — | — | — | — | |
| Vid2SeqPretrain=×2026.04 | 18 | 4 | 4.6 | — | — | — | — | |
| SAILSetting=Weakly-Supervised, Features=CLIP2026.03 | 14.61 | 4.08 | 3.63 | — | 5.42 | 2.94 | — | |
| ILCACMSetting=Weakly-Supervised, Features=CLIP, Implementation=Re-implementation2026.03 | 13.49 | 3.6 | 3.41 | — | 4.75 | 2.59 | — | |
| VTG-LLMPT=✓2026.03 | 13.4 | 3.6 | — | — | — | — | 20.6 | |
| TimeChatPT=✓2026.03 | 11 | 3.4 | — | — | — | — | 19.5 | |
| PWS-DVCSetting=Weakly-Supervised, Features=C3D, Source=Reported in [11]2026.03 | 9.81 | 3.14 | 2.48 | — | — | — | — | |
| WSDECSetting=Weakly-Supervised, Features=C3D, Source=Reported in [11]2026.03 | 8.43 | 2.11 | 1.47 | — | — | — | — | |
| UEDVCTraining paradigm=Fully-Supervised, Pre-trained=676k, Backbone=TSN2024.04 | 8.37 | 3.34 | 2.18 | — | — | — | — | |
| MTFeatures=TSN2023.03 | 6.1 | — | 3.18 | 0.3 | — | — | — | |
| MTBackbone=TSN2023.02 | 3.2 | 6.1 | — | — | — | — | — | |
| BCHRBackbone=C3D2023.02 | — | 3.8 | — | — | — | — | — | |
| E2ESGBackbone=C3D2023.02 | — | 20 | 3.5 | — | — | — | — | |
| ECHRFeatures=TSN2023.03 | — | — | 3.82 | — | — | — | — | |
| SGRFeatures=TSN2023.03 | — | — | 4.35 | — | — | — | — |