TV-VAD: Temporal-frequency assisted video-text alignment for weakly supervised video anomaly detection
Authors:
- Jinjing GuJGJinjing Gu
- Heng ZhangHZHeng Zhang
- Chenggang YangCYChenggang Yang
- Zhengpeng ZhaoZZZhengpeng Zhao
- Yuanyuan PuYPYuanyuan Pu
Preprint
- 15 April 2026
- preprint
- Published by Elsevier in SSRN Electronic Journal
Abstract
Weakly supervised video anomaly detection (WSVAD) aims to identify abnormal events in videos with only coarse video-level annotations. Previous methods primarily focus on modeling anomalies within the spatio-temporal domain of visual modality. However, how to capture global frame-level video features and fully utilize the anchoring role of textual features remains challenging. To address these dilemmas, a novel TV-VAD framework that employs Temporal-frequency assisted Video-text alignment for WSVAD is proposed. Specifically, the temporal-frequency video feature extraction module integrates a Transformer encoder with a dual-branch graph convolutional network to model temporal dynamics, while a Fourier frequency-domain feature recombination module is introduced to capture diverse expressions of anomalies. Subsequently, a video-text alignment block based on Kullback-Leibler divergence aligns the global frame-level video features with fused multi-modal representations, thereby better exploiting the anchoring role of the text modality. Furthermore, a multiclass loss function is employed to reformulate the traditional binary classification task of WSVAD into a multiclass classification task. Extensive experiments demonstrate that TV-VAD achieves state-of-the-art performance on two public benchmarks, achieving an AP of 85.61% on XD-Violence and an AUC of 88.12% on UCF-Crime. Code is available at: https://github.com/jinjinggu00/TV-VAD.Keywords
This publication has 24 references indexed in Scilit:
- Attention-driven pseudo-label self-training for weakly supervised video anomaly detectionPattern Recognition, 2026
- Learning weakly supervised audio-visual violence detection in hyperbolic spaceImage and Vision Computing, 2024
- Delving into CLIP latent space for Video Anomaly RecognitionComputer Vision and Image Understanding, 2024
- SegCFT: Context-aware Fourier Transform for efficient semantic segmentationNeurocomputing, 2024
- Memory-enhanced appearance-motion consistency framework for video anomaly detectionComputer Communications, 2024
- Learning to Prompt for Vision-Language ModelsInternational Journal of Computer Vision, 2022
- Attention-based residual autoencoder for video anomaly detectionApplied Intelligence, 2022
- Contrastive Attention for Video Anomaly DetectionIEEE Transactions on Multimedia, 2021
- CLAWS: Clustering Assisted Weakly Supervised Learning with Normalcy Suppression for Anomalous Event Detection2020
- Not only Look, But Also Listen: Learning Multimodal Violence Detection Under Weak Supervision2020