Tasks SotA Datasets Papers Methods Submit About

Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable Benchmarks All SotA Datasets Papers Methods

Community

Submit Results About

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Models/VideoLights-B-pt

VideoLights-B-pt

Reported on 13 benchmarks across 3 tasks · 1 paper · 2 SOTA

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

Computer Vision11 results

Moment RetrievalonCharades-STA
R@1 IoU=0.3· uses extra data· 2024-12-02
73.33
best: 73.92 (LD-DETR)
SOTA
VideoLights: Feature Refinement and Cross-Task Alignment Transformer for Joint Video Highlight Detection and Moment Retrieval arXiv:2412.01558
Moment RetrievalonCharades-STA
mIoU· uses extra data· 2024-12-02
52.94
best: 53.44 (LD-DETR)
SOTA
VideoLights: Feature Refinement and Cross-Task Alignment Transformer for Joint Video Highlight Detection and Moment Retrieval arXiv:2412.01558
Moment RetrievalonCharades-STA
R@1 IoU=0.5· uses extra data· 2024-12-02
61.96
best: 71.1 (SG-DETR (w/ PT))
VideoLights: Feature Refinement and Cross-Task Alignment Transformer for Joint Video Highlight Detection and Moment Retrieval arXiv:2412.01558
Moment RetrievalonCharades-STA
R@1 IoU=0.7· uses extra data· 2024-12-02
41.05
best: 52.8 (SG-DETR (w/ PT))
VideoLights: Feature Refinement and Cross-Task Alignment Transformer for Joint Video Highlight Detection and Moment Retrieval arXiv:2412.01558
Moment RetrievalonQVHighlights
R@1 IoU=0.5· uses extra data· 2024-12-02
70.36
best: 76.59 (LLaVA-MR)
VideoLights: Feature Refinement and Cross-Task Alignment Transformer for Joint Video Highlight Detection and Moment Retrieval arXiv:2412.01558
Moment RetrievalonQVHighlights
R@1 IoU=0.7· uses extra data· 2024-12-02
55.25
best: 61.48 (LLaVA-MR)
VideoLights: Feature Refinement and Cross-Task Alignment Transformer for Joint Video Highlight Detection and Moment Retrieval arXiv:2412.01558
Moment RetrievalonQVHighlights
mAP· uses extra data· 2024-12-02
47.94
best: 58.8 (SG-DETR (w/ PT))
VideoLights: Feature Refinement and Cross-Task Alignment Transformer for Joint Video Highlight Detection and Moment Retrieval arXiv:2412.01558
Moment RetrievalonQVHighlights
mAP@0.5· uses extra data· 2024-12-02
69.53
best: 76.2 (SG-DETR (w/ PT))
VideoLights: Feature Refinement and Cross-Task Alignment Transformer for Joint Video Highlight Detection and Moment Retrieval arXiv:2412.01558
Moment RetrievalonQVHighlights
mAP@0.75· uses extra data· 2024-12-02
49.17
best: 60.8 (SG-DETR (w/ PT))
VideoLights: Feature Refinement and Cross-Task Alignment Transformer for Joint Video Highlight Detection and Moment Retrieval arXiv:2412.01558
Highlight DetectiononQVHighlights
Hit@1· uses extra data· 2024-12-02
70.56
best: 71.01 (FlashVTG)
VideoLights: Feature Refinement and Cross-Task Alignment Transformer for Joint Video Highlight Detection and Moment Retrieval arXiv:2412.01558
Highlight DetectiononQVHighlights
mAP· uses extra data· 2024-12-02
42.84
best: 44.7 (SG-DETR (w/ PT))
VideoLights: Feature Refinement and Cross-Task Alignment Transformer for Joint Video Highlight Detection and Moment Retrieval arXiv:2412.01558

Methodology2 results

16konQVHighlights
Hit@1· uses extra data· 2024-12-02
70.56
best: 71.01 (FlashVTG)
VideoLights: Feature Refinement and Cross-Task Alignment Transformer for Joint Video Highlight Detection and Moment Retrieval arXiv:2412.01558
16konQVHighlights
mAP· uses extra data· 2024-12-02
42.84
best: 44.7 (SG-DETR (w/ PT))
VideoLights: Feature Refinement and Cross-Task Alignment Transformer for Joint Video Highlight Detection and Moment Retrieval arXiv:2412.01558