UVCOM (w/ PT ASR Captions)

Reported on 5 benchmarks across 1 task · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision5 results

Moment RetrievalonQVHighlights
R@1 IoU=0.5· uses extra data· 2023-11-28
64.53
best: 76.59 (LLaVA-MR)
Bridging the Gap: A Unified Video Comprehension Framework for Moment Retrieval and Highlight Detection arXiv:2311.16464
Moment RetrievalonQVHighlights
R@1 IoU=0.7· uses extra data· 2023-11-28
48.31
best: 61.48 (LLaVA-MR)
Bridging the Gap: A Unified Video Comprehension Framework for Moment Retrieval and Highlight Detection arXiv:2311.16464
Moment RetrievalonQVHighlights
mAP· uses extra data· 2023-11-28
43.8
best: 58.8 (SG-DETR (w/ PT))
Bridging the Gap: A Unified Video Comprehension Framework for Moment Retrieval and Highlight Detection arXiv:2311.16464
Moment RetrievalonQVHighlights
mAP@0.5· uses extra data· 2023-11-28
64.78
best: 76.2 (SG-DETR (w/ PT))
Bridging the Gap: A Unified Video Comprehension Framework for Moment Retrieval and Highlight Detection arXiv:2311.16464
Moment RetrievalonQVHighlights
mAP@0.75· uses extra data· 2023-11-28
43.65
best: 60.8 (SG-DETR (w/ PT))
Bridging the Gap: A Unified Video Comprehension Framework for Moment Retrieval and Highlight Detection arXiv:2311.16464