Learning Prompt-Enhanced Context Features for Weakly-Supervised Video Anomaly Detection
IEEE Transactions on Image Processing · 2024 · 33 · 4923-4936
DOI: 10.1109/tip.2024.3451935 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Image Processing, 2024. Verfasst von Yujiang Pu und 3 weiteren Autor:innen. Die Arbeit wurde bisher 131-mal zitiert (Crossref) und nennt 67 Literaturangaben. Gefördert von: State Key Development Program in 14th Five-Year, National Natural Science Foundation of China, Institute for Guo Qiang, Tsinghua University, and the High-quality and Cutting-Edge Disciplines Construction Project for Universities in Beijing.
Abstract (Originalsprache)
Weakly supervised video anomaly detection aims to locate abnormal activities in untrimmed videos without the need for frame-level supervision. Prior work has utilized graph convolution networks or self-attention mechanisms alongside multiple instance learning (MIL)-based classification loss to model temporal relations and learn discriminative features. However, these approaches are limited in two aspects: 1) Multi-branch parallel architectures, while capturing multi-scale temporal dependencies, inevitably lead to increased parameter and computational costs. 2) The binarized MIL constraint only ensures the interclass separability while neglecting the fine-grained discriminability within anomalous classes. To this end, we introduce a novel WS-VAD framework that focuses on efficient temporal modeling and anomaly innerclass discriminability. We first construct a Temporal Context Aggregation (TCA) module that simultaneously captures local-global dependencies by reusing an attention matrix along with adaptive context fusion. In addition, we propose a Prompt-Enhanced Learning (PEL) module that incorporates semantic priors using knowledge-based prompts to boost the discrimination of visual features while ensuring separability across anomaly subclasses. The proposed components have been validated through extensive experiments, which demonstrate superior performance on three challenging datasets, UCF-Crime, XD-Violence and ShanghaiTech, with fewer parameters and reduced computational effort. Notably, our method can significantly improve the detection accuracy for certain anomaly subclasses and reduced the false alarm rate. Our code is available at: https://github.com/yujiangpu20/PEL4VAD.
Förderung
- State Key Development Program in 14th Five-Year
- National Natural Science Foundation of China
- Institute for Guo Qiang, Tsinghua University, and the High-quality and Cutting-Edge Disciplines Construction Project for Universities in Beijing
Zitieren
APA
Pu, Y., Wu, X., Yang, L., & Wang, S. (2024). Learning Prompt-Enhanced Context Features for Weakly-Supervised Video Anomaly Detection. IEEE Transactions on Image Processing, 33, 4923-4936. https://doi.org/10.1109/tip.2024.3451935
Vancouver
Pu Y, Wu X, Yang L, Wang S. Learning Prompt-Enhanced Context Features for Weakly-Supervised Video Anomaly Detection. IEEE Transactions on Image Processing. 2024;33:4923-4936. doi:10.1109/tip.2024.3451935
BibTeX
@article{Pu2024,
author = {Pu, Yujiang and Wu, Xiaoyu and Yang, Lulu and Wang, Shengjin},
title = {Learning Prompt-Enhanced Context Features for Weakly-Supervised Video Anomaly Detection},
journal = {IEEE Transactions on Image Processing},
year = {2024},
volume = {33},
pages = {4923--4936},
doi = {10.1109/tip.2024.3451935}
}
EUVIMED