Expert-Driven Frequency-Domain Modeling for Video Moment Retrieval
Electronics · 02.11.2026 · 15 (21) · 5080
DOI: 10.3390/electronics15215080 ↗Kurz gefasst
Fachartikel in der Zeitschrift Electronics, online erschienen am 02.11.2026. Verfasst von Jingxi Xie und 5 weiteren Autor:innen. Die Arbeit nennt 63 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China.
Abstract (Originalsprache)
Video moment retrieval (VMR) aims to localize the temporal segment in an untrimmed video that semantically matches a natural-language query. Although recent detection transformer (DETR) approaches have improved this task, they primarily emphasize query-conditioned cross-modal alignment and often underexploit the intrinsic temporal dynamics of videos. In particular, clip-level features contain frequency-domain cues that reveal how visual content changes across time, yet such cues remain underexplored for moment localization. In this paper, we propose FE-DETR, a frequency-domain expert-driven DETR framework for dynamic-aware video moment retrieval. FE-DETR introduces Multi-span Frequency Experts (MFE), which select expert-specific span lengths from frequency-domain scores of a video and perform expert-specific frequency decomposition to capture multi-scale dynamic patterns. The resulting frequency-enhanced representation is fused with regular cross-modal features for query-video alignment. To further improve localization, we design an Expert-targeted Encoder (ETE) with a distinction loss that encourages query-relevant and background clips to be handled by different expert groups, and we incorporate a multi-expert decoder for fine-grained temporal boundary prediction. On TACoS, FE-DETR achieves a three-seed mean of 40.37% Recall@1 at an intersection-over-union (IoU) threshold of 0.5 and a mean IoU of 37.52%. Experiments demonstrate strong performance on TACoS and competitive results on QVHighlights and Charades-STA.
Förderung
- National Natural Science Foundation of China
Zitieren
APA
Xie, J., Ran, R., Cai, X., He, S., Wei, J., & Yang, Y. (2026). Expert-Driven Frequency-Domain Modeling for Video Moment Retrieval. Electronics, 15(21), 5080. https://doi.org/10.3390/electronics15215080
Vancouver
Xie J, Ran R, Cai X, He S, Wei J, Yang Y. Expert-Driven Frequency-Domain Modeling for Video Moment Retrieval. Electronics. 2026;15(21):5080. doi:10.3390/electronics15215080
BibTeX
@article{Xie2026,
author = {Xie, Jingxi and Ran, Ran and Cai, Xiangyi and He, Shiyuan and Wei, Jiwei and Yang, Yang},
title = {Expert-Driven Frequency-Domain Modeling for Video Moment Retrieval},
journal = {Electronics},
year = {2026},
volume = {15},
number = {21},
pages = {5080},
doi = {10.3390/electronics15215080}
}
EUVIMED