Integrating spatial features and dynamically learned temporal features via contrastive learning for video temporal grounding in LLM | OpenReview

Integrating spatial features and dynamically learned temporal features via contrastive learning for video temporal grounding in LLM

Open Webpage

Peifu Wang, Yixiong Liang, Yigang Cen, Lihui Cen, Zhe Qu, Jin Liu, Shichao Kan

Published: 2026, Last Modified: 09 Apr 2026Image Vis. Comput. 2026EveryoneRevisionsBibTeXCC BY-SA 4.0

Loading