Saved in:
Bibliographic Details
Main Authors: Zhou, Hang, Cai, Jiale, Ye, Yuteng, Feng, Yonghui, Gao, Chenxing, Yu, Junqing, Song, Zikai, Yang, Wei
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2412.09026
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913609403072512
author Zhou, Hang
Cai, Jiale
Ye, Yuteng
Feng, Yonghui
Gao, Chenxing
Yu, Junqing
Song, Zikai
Yang, Wei
author_facet Zhou, Hang
Cai, Jiale
Ye, Yuteng
Feng, Yonghui
Gao, Chenxing
Yu, Junqing
Song, Zikai
Yang, Wei
contents A recent endeavor in one class of video anomaly detection is to leverage diffusion models and posit the task as a generation problem, where the diffusion model is trained to recover normal patterns exclusively, thus reporting abnormal patterns as outliers. Yet, existing attempts neglect the various formations of anomaly and predict normal samples at the feature level regardless that abnormal objects in surveillance videos are often relatively small. To address this, a novel patch-based diffusion model is proposed, specifically engineered to capture fine-grained local information. We further observe that anomalies in videos manifest themselves as deviations in both appearance and motion. Therefore, we argue that a comprehensive solution must consider both of these aspects simultaneously to achieve accurate frame prediction. To address this, we introduce innovative motion and appearance conditions that are seamlessly integrated into our patch diffusion model. These conditions are designed to guide the model in generating coherent and contextually appropriate predictions for both semantic content and motion relations. Experimental results in four challenging video anomaly detection datasets empirically substantiate the efficacy of our proposed approach, demonstrating that it consistently outperforms most existing methods in detecting abnormal behaviors.
format Preprint
id arxiv_https___arxiv_org_abs_2412_09026
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Video Anomaly Detection with Motion and Appearance Guided Patch Diffusion Model
Zhou, Hang
Cai, Jiale
Ye, Yuteng
Feng, Yonghui
Gao, Chenxing
Yu, Junqing
Song, Zikai
Yang, Wei
Computer Vision and Pattern Recognition
A recent endeavor in one class of video anomaly detection is to leverage diffusion models and posit the task as a generation problem, where the diffusion model is trained to recover normal patterns exclusively, thus reporting abnormal patterns as outliers. Yet, existing attempts neglect the various formations of anomaly and predict normal samples at the feature level regardless that abnormal objects in surveillance videos are often relatively small. To address this, a novel patch-based diffusion model is proposed, specifically engineered to capture fine-grained local information. We further observe that anomalies in videos manifest themselves as deviations in both appearance and motion. Therefore, we argue that a comprehensive solution must consider both of these aspects simultaneously to achieve accurate frame prediction. To address this, we introduce innovative motion and appearance conditions that are seamlessly integrated into our patch diffusion model. These conditions are designed to guide the model in generating coherent and contextually appropriate predictions for both semantic content and motion relations. Experimental results in four challenging video anomaly detection datasets empirically substantiate the efficacy of our proposed approach, demonstrating that it consistently outperforms most existing methods in detecting abnormal behaviors.
title Video Anomaly Detection with Motion and Appearance Guided Patch Diffusion Model
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2412.09026