Video Grounding on QVHighlights (test)
74.11mAP (IoU=0.5)VideoMind
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| VideoMindSize=2B, Fine-tuned=true2025.03 | 74.11 | 75.42 | 59.35 | 55.15 | 51.6 | — | — | |
| FlashVTGBackbone=InternVideo22024.12 | 72.33 | 70.69 | 53.96 | 53.85 | 52 | — | — | |
| R2-Tunning2024.12 | 69.04 | 68.03 | 49.35 | 47.56 | 46.17 | — | — | |
| R2-TuningFine-tuned=true2025.03 | 69.04 | 68.03 | 49.35 | 47.56 | 46.17 | — | — | |
| FlashVTG2024.12 | 67.99 | 66.08 | 50 | 48.7 | 47.59 | — | — | |
| BAM-DETRPre-training=236K ASR2023.11 | 66.33 | 63.88 | 47.92 | 48.22 | 46.67 | — | — | |
| LLMEPET2024.12 | 65.76 | 66.73 | 49.94 | 43.91 | 44.05 | — | — | |
| BAM-DETRPre-training=none, Audio modality=true2023.11 | 65.61 | 64.07 | 48.12 | 47.51 | 46.91 | — | — | |
| BAM-DETRPre-training=none, Audio modality=false2023.11 | 64.57 | 62.71 | 48.64 | 46.33 | 45.36 | — | — | |
| CG-DETR2024.12 | 64.51 | 65.43 | 48.38 | 42.77 | 42.86 | — | — | |
| UniVTG DBMBlock=DBM2024.03 | 64.37 | 66.65 | 52.19 | 46.68 | 45.18 | 40.18 | 64.77 | |
| UniVTGPre-training=4.2M CLIP2023.11 | 64.06 | 65.43 | 50.06 | 45.02 | 43.63 | — | — | |
| UniVTGFine-tuned=true2025.03 | 64.06 | 65.43 | 50.06 | 45.02 | 43.63 | — | — | |
| TR-DETR2024.12 | 63.98 | 64.66 | 48.96 | 43.73 | 42.62 | — | — | |
| UniVTGBlock=ViM2024.03 | 63.97 | 65.55 | 50.77 | 46.11 | 44.74 | 40.22 | 64.26 | |
| QD-DETRPre-training=236K ASR2023.11 | 63.37 | 63.18 | 45.19 | 40.35 | 39.96 | — | — | |
| UVCOM2024.12 | 63.37 | 63.55 | 47.47 | 42.67 | 43.18 | — | — | |
| QD-NetYear=2023, Setup=fully-supervised2024.03 | 63.15 | 62.32 | 45.61 | 42.05 | 41.46 | — | — | |
| QD-DETRPre-training=none, Audio modality=true2023.11 | 63.04 | 63.06 | 45.1 | 40.1 | 40.19 | — | — | |
| QD-DETRPre-training=none, Audio modality=false2023.11 | 62.52 | 62.4 | 44.98 | 39.88 | 39.86 | — | — | |
| QD-DETR2024.12 | 62.52 | 62.4 | 44.98 | 39.88 | 39.86 | — | — | |
| QD-DETRFine-tuned=true2025.03 | 62.52 | 62.4 | 44.98 | 39.88 | 39.86 | — | — | |
| SFABD2024.12 | 62.38 | — | — | 44.39 | 43.79 | — | — | |
| DenoiseLoc2023.04 | 61.3 | 59.27 | 45.07 | 43.07 | 42.96 | — | — | |
| MH-DETRYear=2022, Setup=fully-supervised2024.03 | 60.75 | 60.05 | 42.48 | 38.13 | 38.38 | — | — | |
| MH-DETRPre-training=none2023.11 | 60.75 | 60.05 | 42.48 | 38.13 | 38.38 | — | — | |
| Moment-DETRPre-training=236K ASR2023.11 | 60.51 | 59.78 | 40.33 | 35.36 | 36.14 | — | — | |
| Moment-DETRFine-tuned=true2025.03 | 60.51 | 59.78 | 40.33 | 35.36 | 36.14 | — | — | |
| EaTR+Pre-training=none2023.11 | 59.95 | 57.98 | 42.41 | 39.29 | 39 | — | — | |
| LLaViLoYear=2023, Setup=fully-supervised2024.03 | 59.72 | 59.23 | 41.42 | — | 36.94 | — | — | |
| UniVTGBlock=Transformer (6-layer, reproduced)2024.03 | 59.09 | 59.87 | 44.32 | 40.25 | 38.48 | 39.22 | 64.71 | |
| UniVTGBlock=Transformer (4-layer)2024.03 | 57.6 | 58.86 | 40.86 | 35.59 | 35.47 | 38.2 | 60.96 | |
| UniVTGPre-training=none2023.11 | 57.6 | 58.86 | 40.86 | 35.59 | 35.47 | — | — | |
| UniVTG2024.12 | 57.6 | 58.86 | 40.86 | 35.59 | 35.47 | — | — | |
| UMTPre-training=236K ASR, Audio modality=true2023.11 | 57.33 | 60.83 | 43.26 | 39.12 | 38.08 | — | — | |
| UMTFine-tuned=true2025.03 | 57.33 | 60.83 | 43.26 | 39.12 | 38.08 | — | — | |
| Moment-DETR2023.04 | 54.82 | 52.89 | 33.02 | 29.4 | 30.73 | — | — | |
| MDETRBlock=-2024.03 | 54.82 | 52.89 | 33.02 | 29.4 | 30.73 | 35.69 | 55.6 | |
| Moment-DETRYear=2021, Setup=fully-supervised2024.03 | 54.82 | 52.89 | 33.02 | 29.4 | 30.73 | — | — | |
| Moment-DETRPre-training=none2023.11 | 54.82 | 52.89 | 33.02 | 29.4 | 30.73 | — | — | |
| M-DETR2024.12 | 54.82 | 52.89 | 33.02 | 29.17 | 30.73 | — | — | |
| MomentDiffFine-tuned=true2025.03 | 54.57 | 58.21 | 41.48 | 37.21 | 36.84 | — | — | |
| VTG-GPTYear=2023, Setup=zero-shot2024.03 | 54.13 | 53.81 | 38.13 | 29.24 | 30.5 | — | — | |
| Moment DiffPre-training=none2023.11 | 54.02 | 57.42 | 39.66 | 35.73 | 35.95 | — | — | |
| MomentDiff2024.12 | 54.02 | 57.42 | 39.66 | 35.73 | 35.95 | — | — | |
| UMT2023.04 | 53.83 | 56.23 | 41.18 | 37.01 | 36.12 | — | — | |
| UMTBlock=-2024.03 | 53.83 | 56.23 | 41.18 | 37.01 | 36.12 | 38.18 | 59.99 | |
| UMTYear=2023, Setup=fully-supervised2024.03 | 53.83 | 56.23 | 41.18 | 37.01 | 36.12 | — | — | |
| UMTPre-training=none, Audio modality=true, Reproduced=true2023.11 | 53.83 | 56.23 | 41.18 | 37.01 | 36.12 | — | — | |
| UMT2024.12 | 53.83 | 56.23 | 41.18 | 37.01 | 36.12 | — | — | |
| XML+2023.04 | 47.89 | 46.69 | 33.46 | 34.67 | 34.9 | — | — | |
| XMLPre-training=none2023.11 | 47.89 | 46.69 | 33.46 | 34.67 | 34.9 | — | — | |
| XML+Fine-tuned=true2025.03 | 47.89 | 46.69 | 33.46 | 34.67 | 34.9 | — | — | |
| XML2023.04 | 44.63 | 41.83 | 30.35 | 31.73 | 32.14 | — | — | |
| XMLFine-tuned=true2025.03 | 44.63 | 41.83 | 30.35 | 31.73 | 32.14 | — | — | |
| UniVTGYear=2023, Setup=zero-shot2024.03 | 27.42 | 25.16 | 8.95 | 7.64 | 10.87 | — | — | |
| MCN2023.04 | 24.94 | 11.41 | 2.72 | 8.22 | 10.67 | — | — | |
| MCNPre-training=none2023.11 | 24.94 | 11.41 | 2.72 | 8.22 | 10.67 | — | — | |
| CPIYear=2022, Setup=weakly supervised2024.03 | 23.74 | 32.26 | 11.81 | 8.25 | — | — | — | |
| CAL2023.04 | 23.4 | 25.49 | 11.54 | 7.65 | 9.89 | — | — | |
| CALPre-training=none2023.11 | 23.4 | 25.49 | 11.54 | 7.65 | 9.89 | — | — | |
| CPLYear=2022, Setup=weakly supervised2024.03 | 22.77 | 30.72 | 10.75 | 7.48 | — | — | — | |
| CLIPPre-training=none2023.11 | 18.11 | 16.88 | 5.19 | 7 | 7.67 | — | — | |
| CNMYear=2022, Setup=weakly supervised2024.03 | 11.78 | 14.11 | 3.97 | 2.12 | — | — | — |