Coarse To Fine: Video Retrieval Before Moment Localization
Zijian Gao, Huanyu Liu, Jingyu Liu . Arxiv 2021 – 1 citation
[Paper]
The current state-of-the-art methods for video corpus moment retrieval (VCMR) often use similarity-based feature alignment approach for the sake of convenience and speed. However, late fusion methods like cosine similarity alignment are unable to make full use of the information from both query texts and videos. In this paper, we combine feature alignment with feature fusion to promote the performance on VCMR.