MS-DETR improves moment retrieval and highlight detection by disentangling motion and semantic video features, sharing task information between the two tasks, and training on generated auxiliary captions.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
MS-DETR: Towards Effective Video Moment Retrieval and Highlight Detection by Joint Motion-Semantic Learning
MS-DETR improves moment retrieval and highlight detection by disentangling motion and semantic video features, sharing task information between the two tasks, and training on generated auxiliary captions.