UMT: Unified Multi-modal Transformers for Joint Video Moment Retrieval and Highlight Detection

Finding relevant moments and highlights in videos according to natural language queries is a natural and highly valuable common need in the current video content explosion era. Nevertheless, jointly conducting moment retrieval and highlight detection is an emerging research topic, even though its component problems and some related tasks have already been studied for a while. In this paper, we present the first unified framework, named Unified Multi-modal Transformers (UMT), capable of realizing such joint optimization while can also be easily degenerated for solving individual problems. As far as we are aware, this is the first scheme to integrate multi-modal (visual-audio) learning for either joint optimization or the individual moment retrieval task, and tackles moment retrieval as a keypoint detection problem using a novel query generator and query decoder. Extensive comparisons with existing methods and ablation studies on QVHighlights, Charades-STA, YouTube Highlights, and TVSum datasets demonstrate the effectiveness, superiority, and flexibility of the proposed method under various settings. Source code and pre-trained models are available at

PDF Abstract CVPR 2022 PDF CVPR 2022 Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Moment Retrieval Charades-STA UMT (VA) R@1 IoU=0.5 48.31 # 7
R@1 IoU=0.7 29.25 # 6
R@5 IoU=0.5 88.79 # 2
R@5 IoU=0.7 56.08 # 3
Moment Retrieval Charades-STA UMT (VO) R@1 IoU=0.5 49.35 # 6
R@1 IoU=0.7 26.16 # 8
R@5 IoU=0.5 89.41 # 1
R@5 IoU=0.7 54.95 # 5
Moment Retrieval QVHighlights UMT mAP 36.12 # 8
Moment Retrieval QVHighlights UMT (w. PT) mAP 38.08 # 6
Highlight Detection QVHighlights UMT (w. PT) mAP 39.12 # 2
Highlight Detection QVHighlights UMT mAP 38.18 # 7
Highlight Detection TvSum UMT mAP 83.1 # 3
Highlight Detection YouTube Highlights UMT mAP 74.9 # 1


No methods listed for this paper. Add relevant methods here