Moment Retrieval on Charades-STA (test)
70.32R@0.5FlashVTG
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| FlashVTGBackbone=InternVideo2-6B2024.12 | 70.32 | — | 49.87 | — | — | — | — | — | |
| VideoLights-Bpre-training=true, features=BLIP2024.12 | 61.96 | 73.33 | 41.05 | 52.94 | — | — | — | — | |
| LA-DETRFeatures=SlowFast/CLIP2024.12 | 61.45 | — | 40.22 | — | — | — | — | — | |
| VideoLights-Bpre-training=false, features=BLIP2024.12 | 60.3 | 71.72 | 37.23 | 51.25 | — | — | — | — | |
| UniVTGyear=2023, venue=ICCV, pre-training=true2024.12 | 60.19 | 72.63 | 38.55 | 52.17 | — | — | — | — | |
| FlashVTGFeatures=SlowFast/CLIP2024.12 | 60.11 | — | 38.01 | — | — | — | — | — | |
| VideoLightspre-training=true, features=standard2024.12 | 60.11 | 72.26 | 37.8 | 51.44 | — | — | — | — | |
| FlashVTGBackbone=SlowFast R-50 + CLIP-B/322024.12 | 60.11 | — | 38.01 | — | — | — | — | — | |
| BAM-DETRFeatures=SlowFast/CLIP2024.12 | 59.83 | — | 39.83 | — | — | — | — | — | |
| R2-TuningFeature Extractors=CLIP-B/322024.03 | 59.78 | 70.91 | 37.02 | 50.86 | — | — | — | — | |
| UVCOMFeatures=SlowFast/CLIP2024.12 | 59.25 | — | 36.64 | — | — | — | — | — | |
| UVCOMFeatures=SlowFast + CLIP2023.11 | 59.25 | — | 36.64 | — | — | — | — | — | |
| CG-DETRBackbone=SlowFast R-50 + CLIP-B/322024.12 | 58.44 | — | 36.34 | — | — | — | — | — | |
| CG-DETRBackbone=Slowfast and CLIP2023.11 | 58.4 | 70.4 | 36.3 | 50.1 | — | — | — | — | |
| VideoLightspre-training=false, features=standard2024.12 | 58.04 | 70.67 | 36.88 | 50.2 | — | — | — | — | |
| UniVTGFeatures=SlowFast/CLIP2024.12 | 58.01 | — | 35.65 | — | — | — | — | — | |
| UniVTGFeatures=SlowFast + CLIP2023.11 | 58.01 | — | 35.65 | — | — | — | — | — | |
| UniVTGFeat.=SF+C, Protocol=S2@202023.08 | 58.01 | — | 35.65 | — | — | — | — | — | |
| UniVTGFeature Extractors=CLIP-B/32 and SlowFast R-502024.03 | 58.01 | 70.81 | 35.65 | 50.1 | — | — | — | — | |
| UniVTGyear=2023, venue=ICCV2024.12 | 58.01 | 70.81 | 35.65 | 50.1 | — | — | — | — | |
| UniVTGBackbone=SlowFast R-50 + CLIP-B/322024.12 | 58.01 | — | 35.65 | — | — | — | — | — | |
| UniVTGBackbone=Slowfast and CLIP2023.11 | 58 | 70.8 | 35.7 | 50.1 | — | — | — | — | |
| TR-DETRFeatures=SlowFast/CLIP2024.12 | 57.61 | — | 33.52 | — | — | — | — | — | |
| TR-DETRyear=2024, venue=AAAI2024.12 | 57.61 | — | 33.52 | — | — | — | — | — | |
| TR-DETRBackbone=SlowFast R-50 + CLIP-B/322024.12 | 57.61 | — | 33.52 | — | — | — | — | — | |
| GroundVTS-QBackbone=Qwen2.5VL-7B2026.04 | 57.5 | 71.5 | 34.2 | 50.1 | — | — | — | — | |
| QD-DETRFeatures=SlowFast/CLIP2024.12 | 57.31 | — | 32.55 | — | — | — | — | — | |
| QD-DETRfeat=SF+C, Source Modality=Video2023.03 | 57.31 | — | 32.55 | — | — | — | — | — | |
| QD-DETRFeatures=SlowFast + CLIP2023.11 | 57.31 | — | 32.55 | — | — | — | — | — | |
| QD-DETRFeat.=SF+C, Protocol=S2@202023.08 | 57.31 | — | 32.55 | — | — | — | — | — | |
| QD-DETRyear=2023, venue=CVPR2024.12 | 57.31 | — | 32.55 | — | — | — | — | — | |
| QD-DETRBackbone=SlowFast R-50 + CLIP-B/322024.12 | 57.31 | — | 32.55 | — | — | — | — | — | |
| QD-DETRBackbone=Slowfast and CLIP2023.11 | 57.3 | — | 32.6 | — | — | — | — | — | |
| UVCOMaudio modality=true2023.11 | 56.69 | — | 34.76 | — | — | — | — | — | |
| TaskWeaveBackbone=VGG2024.04 | 56.51 | — | 33.66 | — | — | — | — | — | |
| LA-DETRFeatures=VGG/GloVe2024.12 | 56.16 | — | 36.1 | — | — | — | — | — | |
| LLaViLoBackbone=Slowfast and CLIP2023.11 | 55.7 | — | 33.4 | — | — | — | — | — | |
| Moment-DETRPre-training=10K HowTo100M videos2021.07 | 55.65 | — | 34.17 | — | — | — | — | — | |
| MomentDiffType=SF+C, C2023.07 | 55.57 | — | 32.42 | — | 61.07 | 32.51 | 32.85 | — | |
| QD-DETRfeat=VGG, Source Modality=Video + Audio2023.03 | 55.51 | — | 34.17 | — | — | — | — | — | |
| QD-DETRaudio modality=true2023.11 | 55.51 | — | 34.17 | — | — | — | — | — | |
| MH-DETRFeat.=VGG, Protocol=S2@202023.08 | 55.47 | — | 32.41 | — | — | — | — | — | |
| MomentDiffType=SF+C, Glove2023.07 | 55.42 | — | 32.17 | — | 60.93 | 32.47 | 32.59 | — | |
| CG-DETRBackbone=VGG2024.12 | 55.22 | — | 34.19 | — | — | — | — | — | |
| CG-DETRBackbone=VGG, Audio modality=false2023.11 | 55.2 | — | 34.2 | — | — | — | — | — | |
| UVCOMFeatures=VGG/GloVe2024.12 | 54.57 | — | 34.13 | — | — | — | — | — | |
| UVCOMaudio modality=false2023.11 | 54.57 | — | 34.13 | — | — | — | — | — | |
| FlashVTGFeatures=VGG/GloVe2024.12 | 54.25 | — | 37.42 | — | — | — | — | — | |
| FlashVTGBackbone=VGG2024.12 | 54.25 | — | 37.42 | — | — | — | — | — | |
| MomentDiffType=C3D, Glove2023.07 | 53.79 | — | 30.18 | — | 59.32 | 29.85 | 31.89 | — | |
| M-DETRFeatures=SlowFast/CLIP2024.12 | 53.63 | — | 31.37 | — | — | — | — | — | |
| Moment-DETR2021.07 | 53.63 | — | 31.37 | — | — | — | — | — | |
| M-DETRfeat=SF+C, Source Modality=Video2023.03 | 53.63 | — | 31.37 | — | — | — | — | — | |
| M-DETRFeatures=SlowFast + CLIP2023.11 | 53.63 | — | 31.37 | — | — | — | — | — | |
| M-DETRFeat.=SF+C, Protocol=S2@202023.08 | 53.63 | — | 31.37 | — | — | — | — | — | |
| TR-DETRFeatures=VGG/GloVe2024.12 | 53.47 | — | 30.81 | — | — | — | — | — | |
| TR-DETRBackbone=VGG2024.12 | 53.47 | — | 30.81 | — | — | — | — | — | |
| TaskWeaveBackbone=I3D2024.04 | 53.36 | — | 31.4 | — | — | — | — | — | |
| MomentDETRType=SF+C, C, re-implemented=true2023.07 | 53.22 | — | 30.87 | — | 58.86 | 26.43 | 30.43 | — | |
| QD-DETRBackbone=VGG, Audio modality=false2023.11 | 52.8 | — | 31.1 | — | — | — | — | — | |
| QD-DETRFeatures=VGG/GloVe2024.12 | 52.77 | — | 31.13 | — | — | — | — | — | |
| QD-DETRfeat=VGG, Source Modality=Video2023.03 | 52.77 | — | 31.13 | — | — | — | — | — | |
| QD-DETRaudio modality=false2023.11 | 52.77 | — | 31.13 | — | — | — | — | — | |
| QD-DETRFeat.=VGG, Protocol=S2@202023.08 | 52.77 | — | 31.13 | — | — | — | — | — | |
| QD-DETRBackbone=VGG2024.04 | 52.77 | — | 31.13 | — | — | — | — | — | |
| QD-DETRBackbone=VGG2024.12 | 52.77 | — | 31.13 | — | — | — | — | — | |
| MomentDiffType=VGG+A, Glove, audio data=true2023.07 | 52.62 | — | 29.93 | — | 60.69 | 29.74 | 31.81 | — | |
| M-DETRBackbone=Slowfast and CLIP2023.11 | 52.1 | 65.8 | 30.6 | 45.5 | — | — | — | — | |
| Moment-DETRFeature Extractors=CLIP-B/322024.03 | 52.07 | 65.83 | 30.59 | 45.54 | — | — | — | — | |
| Moment-DETRyear=2021, venue=NIPS2024.12 | 52.07 | 65.83 | 30.59 | 45.54 | — | — | — | — | |
| Moment-DETRBackbone=SlowFast R-50 + CLIP-B/322024.12 | 52.07 | — | 30.59 | — | — | — | — | — | |
| MomentDiffType=VGG, Glove2023.07 | 51.94 | — | 28.25 | — | 59.86 | 29.11 | 31.66 | — | |
| QD-DETRfeat=C3D, Source Modality=Video2023.03 | 50.67 | — | 31.02 | — | — | — | — | — | |
| QD-DETRBackbone=I3D2024.04 | 50.67 | — | 31.02 | — | — | — | — | — | |
| MomentDETRType=VGG, Glove, re-implemented=true2023.07 | 50.54 | — | 28.01 | — | 57.39 | 25.62 | 29.87 | — | |
| MomentDETRType=C3D, Glove, re-implemented=true2023.07 | 50.49 | — | 29.95 | — | 56.27 | 26.08 | 29.92 | — | |
| IVG-DCL2021.07 | 50.24 | — | 32.88 | — | — | — | — | — | |
| IVG-DCLFeat.=SF+C, Protocol=S2@202023.08 | 50.24 | — | 32.88 | — | — | — | — | — | |
| TFVTG(GPT-4T)Fine-tuning=true2025.01 | 49.9 | — | — | 44.5 | — | — | — | — | |
| UMTFeat.=VGG+Opti., Protocol=S2@202023.08 | 49.35 | — | 26.16 | — | — | — | — | — | |
| VSLNetType=C3D, Glove, re-implemented=true2023.07 | 48.67 | — | 30.33 | — | 56.88 | 25.79 | 30.16 | — | |
| UMTType=VGG+A, Glove, audio data=true2023.07 | 48.44 | — | 29.76 | — | 58.03 | 27.46 | 30.37 | — | |
| UMTfeat=VGG, Source Modality=Video + Audio2023.03 | 48.31 | — | 29.25 | — | — | — | — | — | |
| UMTaudio modality=true2023.11 | 48.31 | — | 29.25 | — | — | — | — | — | |
| UMTBackbone=VGG, Audio modality=true2024.04 | 48.31 | — | 29.25 | — | — | — | — | — | |
| UMT+Backbone=VGG, Audio features=true2024.12 | 48.31 | — | 29.25 | — | — | — | — | — | |
| UMTBackbone=VGG, Audio modality=true2023.11 | 48.3 | — | 29.3 | — | — | — | — | — | |
| CBLNType=VGG, Glove2023.07 | 47.94 | — | 28.22 | — | — | — | — | — | |
| DCMType=VGG, Glove2023.07 | 47.8 | — | 28 | — | — | — | — | — | |
| VSLNet2021.07 | 47.31 | — | 30.19 | — | — | — | — | — | |
| VSLNetfeat=C3D, Source Modality=Video2023.03 | 47.31 | — | 30.19 | — | — | — | — | — | |
| CALFeat.=SF+C, Protocol=S2@202023.08 | 47.31 | — | 30.19 | — | — | — | — | — | |
| VSLNetFeat.=SF+C, Protocol=S2@202023.08 | 47.31 | — | 30.19 | — | — | — | — | — | |
| VSLNetBackbone=I3D2024.04 | 47.31 | — | 30.19 | — | — | — | — | — | |
| MMNType=VGG, Glove, re-implemented=true2023.07 | 46.93 | — | 27.07 | — | 58.85 | 28.16 | 31.58 | — | |
| SSRNBackbone=VGG, Audio modality=false2023.11 | 46.7 | — | 28 | — | — | — | — | — | |
| MANfeat=C3D, Source Modality=Video2023.03 | 46.53 | — | 22.72 | — | — | — | — | — | |
| MANBackbone=I3D2024.04 | 46.53 | — | 22.72 | — | — | — | — | — | |
| 2D-TANFeatures=SlowFast/CLIP2024.12 | 46.02 | — | 27.5 | — | — | — | — | — | |
| 2D TANFeatures=SlowFast + CLIP2023.11 | 46.02 | — | 27.5 | — | — | — | — | — |