RelVid: Relational Learning with Vision-Language Models for Weakly Video Anomaly Detection

作者
Jingxin Wang,Guohan Li,Jiaqi Liu,Zhengyi Xu,Xinrong Chen,Jianming Wei
出处
期刊:Sensors [Multidisciplinary Digital Publishing Institute]
卷期号:25 (7): 2037-2037 被引量:12
标识
DOI:10.3390/s25072037
摘要

Weakly supervised video anomaly detection aims to identify abnormal events in video sequences without requiring frame-level supervision, which is a challenging task in computer vision. Traditional methods typically rely on low-level visual features with weak supervision from a single backbone branch, which often struggles to capture the distinctive characteristics of different categories. This limitation reduces their adaptability to real-world scenarios. In real-world situations, the boundary between normal and abnormal events is often unclear and context-dependent. For example, running on a track may be considered normal, but running on a busy road could be deemed abnormal. To address these challenges, RelVid is introduced as a novel framework that improves anomaly detection by expanding the relative feature gap between classes extracted from a single backbone branch. The key innovation of RelVid lies in the integration of auxiliary tasks, which guide the model to learn more discriminative features, significantly boosting the model's performance. These auxiliary tasks-including text-based anomaly detection and feature reconstruction learning-act as additional supervision, helping the model capture subtle differences and anomalies that are often difficult to detect in weakly supervised settings. In addition, RelVid incorporates two other components, which include class activation feature learning for improved feature discrimination and a temporal attention module for capturing sequential dependencies. This approach enhances the model's robustness and accuracy, enabling it to better handle complex and ambiguous scenarios. Evaluations on two widely used benchmark datasets, UCF-Crime and XD-Violence, demonstrate the effectiveness of RelVid. Compared to state-of-the-art methods, RelVid achieves superior performance in both detection accuracy and robustness.

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
刚刚
lucky发布了新的文献求助10
刚刚
yiyi完成签到,获得积分10
1秒前
大气的妙之完成签到,获得积分10
1秒前
CipherSage的应助被Code_king采纳,获得10
1秒前
海鹰完成签到,获得积分10
1秒前
1秒前
馒头完成签到,获得积分10
1秒前
桐桐的应助被Code_king采纳,获得10
1秒前
搜集达人的应助被Code_king采纳,获得10
1秒前
科目三的应助被andars0828采纳,获得10
1秒前
笑点低的靖仇完成签到,获得积分10
2秒前
2秒前
SciGPT的应助被Code_king采纳,获得10
2秒前
可爱的函函的应助被Code_king采纳,获得10
2秒前
2秒前
香蕉觅云的应助被Code_king采纳,获得10
2秒前
无花果的应助被Code_king采纳,获得10
2秒前
一一完成签到 ,获得积分10
3秒前
赘婿的应助被谭智航采纳,获得10
3秒前
Owen的应助被Code_king采纳,获得10
3秒前
wjx发布了新的文献求助10
3秒前
余香的应助被Code_king采纳,获得10
3秒前
3秒前
一只王火火完成签到,获得积分10
3秒前
3秒前
3秒前
标致小土豆完成签到 ,获得积分10
3秒前
Flora314发布了新的文献求助10
4秒前
刘天煜发布了新的文献求助10
4秒前
4秒前
TFboy完成签到,获得积分10
4秒前
4秒前
4秒前
4秒前
wang完成签到,获得积分10
5秒前
传奇3的应助被乐迪采纳,获得10
5秒前
JoanJin发布了新的文献求助10
5秒前
枫泾发布了新的文献求助10
5秒前
积极电脑发布了新的文献求助10
5秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 1700
奈特基础组织学:配套组织病理学(第4版)(Netter's Essential Histology With Correlated Histopathology) 800
John McDowells Bildungstheorie 800
八旗制度の研究 600
plantTFDB 508
Resource Recovery Handbook 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8176174
求助须知:如何正确求助?哪些是违规求助? 9567656
关于积分的说明 21793895
捐赠科研通 7770887
什么是DOI,文献DOI怎么找? 3431991
关于科研通互助平台的介绍 2556313
邀请新用户注册赠送积分活动 2477540