Coarse to Fine: Video Retrieval before Moment Localization
This work addresses video retrieval challenges for researchers and practitioners, but it appears incremental as it builds on existing methods by integrating feature fusion.
The paper tackles the problem of video corpus moment retrieval by combining feature alignment with feature fusion to improve performance, addressing limitations of late fusion methods like cosine similarity alignment.
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.