BFSTAL: Bidirectional Feature Splitting With Cross-Layer Fusion for Temporal Action Localization
IEEE Transactions on Circuits and Systems for Video Technology · 2025-12 · 35 (12) · 12707-12718
DOI: 10.1109/tcsvt.2025.3588710 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Circuits and Systems for Video Technology, 2025-12. Verfasst von Jinglin Xu und 3 weiteren Autor:innen. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 76 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Beijing Natural Science Foundation, Young Elite Scientists Sponsorship Program by the China Association for Science and Technology.
Abstract (Originalsprache)
Temporal Action Localization (TAL) aims to identify the boundaries of actions and their corresponding categories in untrimmed videos. Most existing methods simultaneously process past and future information, neglecting the inherently sequential nature of action occurrence. This confused treatment of past and future information hinders the model’s ability to understand action procedures effectively. To address these issues, we propose Bidirectional Feature Splitting with Cross-Layer Fusion for Temporal Action Localization (BFSTAL), a new bidirectional feature-splitting approach based on Mamba for the TAL task, composed of two core parts, Decomposed Bidirectionally Hybrid (DBH) and Cross-Layer Fusion Detection (CLFD), which explicitly enhances the model’s capacity to understand action procedures, especially to localize temporal boundaries of actions. Specifically, we introduce the Decomposed Bidirectionally Hybrid (DBH) component, which splits video features at a given timestamp into forward features (past information) and backward features (future information). DBH integrates three key modules: Bidirectional Multi-Head Self-Attention (Bi-MHSA), Bidirectional State Space Model (Bi-SSM), and Bidirectional Convolution (Bi-CONV). DBH effectively captures long-range dependencies by combining state-space modeling, attention mechanisms, and convolutional networks while improving spatial-temporal awareness. Furthermore, we propose Cross-Layer Fusion Detection (CLFD), which aggregates multi-scale features from different pyramid levels, enhancing contextual understanding and temporal action localization precision. Extensive experiments demonstrate that BFSTAL outperforms other methods on four widely used TAL benchmarks: THUMOS14, EPIC-KITCHENS 100, Charades, and MultiTHUMOS.
Förderung
- National Natural Science Foundation of China
- Beijing Natural Science Foundation
- Young Elite Scientists Sponsorship Program by the China Association for Science and Technology
Zitieren
APA
Xu, J., Zhang, Y., Zhou, W., & Liu, H. (2025). BFSTAL: Bidirectional Feature Splitting With Cross-Layer Fusion for Temporal Action Localization. IEEE Transactions on Circuits and Systems for Video Technology, 35(12), 12707-12718. https://doi.org/10.1109/tcsvt.2025.3588710
Vancouver
Xu J, Zhang Y, Zhou W, Liu H. BFSTAL: Bidirectional Feature Splitting With Cross-Layer Fusion for Temporal Action Localization. IEEE Transactions on Circuits and Systems for Video Technology. 2025;35(12):12707-12718. doi:10.1109/tcsvt.2025.3588710
BibTeX
@article{Xu2025,
author = {Xu, Jinglin and Zhang, Yaqi and Zhou, Wenhao and Liu, Hongmin},
title = {BFSTAL: Bidirectional Feature Splitting With Cross-Layer Fusion for Temporal Action Localization},
journal = {IEEE Transactions on Circuits and Systems for Video Technology},
year = {2025},
volume = {35},
number = {12},
pages = {12707--12718},
doi = {10.1109/tcsvt.2025.3588710}
}
EUVIMED