Enregistré dans:
Détails bibliographiques
Auteurs principaux: Lee, Junhee, Bang, ChaeBeen, Kim, MyoungChul, Cho, MyeongAh
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:https://arxiv.org/abs/2511.13204
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866918264944197632
author Lee, Junhee
Bang, ChaeBeen
Kim, MyoungChul
Cho, MyeongAh
author_facet Lee, Junhee
Bang, ChaeBeen
Kim, MyoungChul
Cho, MyeongAh
contents Weakly-Supervised Video Anomaly Detection aims to identify anomalous events using only video-level labels, balancing annotation efficiency with practical applicability. However, existing methods often oversimplify the anomaly space by treating all abnormal events as a single category, overlooking the diverse semantic and temporal characteristics intrinsic to real-world anomalies. Inspired by how humans perceive anomalies, by jointly interpreting temporal motion patterns and semantic structures underlying different anomaly types, we propose RefineVAD, a novel framework that mimics this dual-process reasoning. Our framework integrates two core modules. The first, Motion-aware Temporal Attention and Recalibration (MoTAR), estimates motion salience and dynamically adjusts temporal focus via shift-based attention and global Transformer-based modeling. The second, Category-Oriented Refinement (CORE), injects soft anomaly category priors into the representation space by aligning segment-level features with learnable category prototypes through cross-attention. By jointly leveraging temporal dynamics and semantic structure, explicitly models both "how" motion evolves and "what" semantic category it resembles. Extensive experiments on WVAD benchmark validate the effectiveness of RefineVAD and highlight the importance of integrating semantic context to guide feature refinement toward anomaly-relevant patterns.
format Preprint
id arxiv_https___arxiv_org_abs_2511_13204
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle RefineVAD: Semantic-Guided Feature Recalibration for Weakly Supervised Video Anomaly Detection
Lee, Junhee
Bang, ChaeBeen
Kim, MyoungChul
Cho, MyeongAh
Computer Vision and Pattern Recognition
Weakly-Supervised Video Anomaly Detection aims to identify anomalous events using only video-level labels, balancing annotation efficiency with practical applicability. However, existing methods often oversimplify the anomaly space by treating all abnormal events as a single category, overlooking the diverse semantic and temporal characteristics intrinsic to real-world anomalies. Inspired by how humans perceive anomalies, by jointly interpreting temporal motion patterns and semantic structures underlying different anomaly types, we propose RefineVAD, a novel framework that mimics this dual-process reasoning. Our framework integrates two core modules. The first, Motion-aware Temporal Attention and Recalibration (MoTAR), estimates motion salience and dynamically adjusts temporal focus via shift-based attention and global Transformer-based modeling. The second, Category-Oriented Refinement (CORE), injects soft anomaly category priors into the representation space by aligning segment-level features with learnable category prototypes through cross-attention. By jointly leveraging temporal dynamics and semantic structure, explicitly models both "how" motion evolves and "what" semantic category it resembles. Extensive experiments on WVAD benchmark validate the effectiveness of RefineVAD and highlight the importance of integrating semantic context to guide feature refinement toward anomaly-relevant patterns.
title RefineVAD: Semantic-Guided Feature Recalibration for Weakly Supervised Video Anomaly Detection
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2511.13204