Learning Multimodal Deep Representations for Crowd Anomaly Event Detection

Huang Shaonian; Huang Dongjun; Zhou Xinmin

首页> 外文期刊>Mathematical Problems in Engineering >Learning Multimodal Deep Representations for Crowd Anomaly Event Detection

【24h】

Learning Multimodal Deep Representations for Crowd Anomaly Event Detection

机译：学习用于人群异常事件检测的多模式深度表示

获取原文

获取原文并翻译 | 示例

掌桥外文数据库（机构版） >>

开具论文收录证明 >>

文献代查 >>

页面导航

摘要
著录项
相似文献
相关主题

摘要

Anomaly event detection in crowd scenes is extremely important; however, the majority of existing studiesmerely use hand-crafted features to detect anomalies. In this study, a novel unsupervised deep learning framework is proposed to detect anomaly events in crowded scenes. Specifically, low-level visual features, energy features, andmotionmap features are simultaneously extracted based on spatiotemporal energy measurements. Three convolutional restricted Boltzmann machines are trained to model the mid-level feature representation of normal patterns. Then a multimodal fusion scheme is utilized to learn the deep representation of crowd patterns. Based on the learned deep representation, a one-class support vector machinemodel is used to detect anomaly events. The proposed method is evaluated using two available public datasets and compared with state-of-the-art methods. The experimental results show its competitive performance for anomaly event detection in video surveillance.

机译：人群场景中的异常事件检测非常重要。但是，大多数现有研究仅使用手工制作的特征来检测异常。在这项研究中，提出了一种新颖的无监督深度学习框架来检测拥挤场景中的异常事件。具体地，基于时空能量测量同时提取低级视觉特征，能量特征和运动图特征。训练了三台卷积受限的Boltzmann机器，以对正常模式的中层特征表示进行建模。然后，采用多峰融合方案来学习人群模式的深度表示。基于学习到的深度表示，一类支持向量机模型用于检测异常事件。使用两个可用的公共数据集对提出的方法进行了评估，并与最新方法进行了比较。实验结果表明，它在视频监视中的异常事件检测方面具有竞争力。

著录项

来源
《Mathematical Problems in Engineering》 |2018年第2期|6323942.1-6323942.13|共13页
作者
Huang Shaonian; Huang Dongjun; Zhou Xinmin;
展开▼
作者单位

Cent S Univ, Sch Informat Sci & Engn, Changsha 410083, Hunan, Peoples R China;

Cent S Univ, Sch Informat Sci & Engn, Changsha 410083, Hunan, Peoples R China;

Hunan Univ Commerce, Sch Comp & Informat Engn, Key Lab Hunan Prov New Retail Virtual Real Techno, Changsha 410205, Hunan, Peoples R China;

展开▼
收录信息
原文格式 PDF
正文语种 eng
中图分类
关键词

相似文献

外文文献
中文文献
专利

1. Learning Multimodal Deep Representations for Crowd Anomaly Event Detection [J] . Shaonian Huang, Dongjun Huang, Xinmin Zhou Mathematical Problems in Engineering: Theory, Methods and Applications . 2018,第6期

机译：学习用于人群异常事件检测的多模式深度表示
2. Learning deep event models for crowd anomaly detection [J] . Feng Yachuang, Yuan Yuan, Lu Xiaoqiang Neurocomputing . 2017,第JANa5期

机译：学习深度事件模型以检测人群异常
3. Revisiting crowd behaviour analysis through deep learning: Taxonomy, anomaly detection, crowd emotions, datasets, opportunities and prospects [J] . Luque Sanchez Francisco, Hupont Isabelle, Tabik Siham, Information Fusion . 2020,第1期

机译：通过深入学习重新审视人群行为分析：分类，异常检测，人群情绪，数据集，机会和前景
4. Application of Deep Learning for Crowd Anomaly Detection from Surveillance Videos [C] . Karishma Pawar, Vahida Attar International Conference on Cloud Computing, Data Science and Engineering . 2021

机译：深度学习对人群异常检测监测视频的应用
5. Anomaly Compressor Audio Data Detection and Classification using Deep Learning Techniques [D] . Zhang, Xulong. 2021

机译：异常压缩机音频数据检测和使用深度学习技术的分类
6. Multimodal deep representation learning for protein interaction identification and protein family classification [O] . Da Zhang, Mansur Kabuka 2019

机译：用于蛋白质相互作用鉴定和蛋白质家族分类的多模式深度表示学习
7. Comparison of deep learning-based methods in multimodal anomaly detection: A case study in human–robot collaboration [O] . Lin Yang, Wu Yan, Hongmin Wu 2021

机译：基于深度学习的多模式异常检测方法的比较 - 以人 - 机器人协作为例

Learning Multimodal Deep Representations for Crowd Anomaly Event Detection

摘要

著录项

相似文献

相关主题

期刊订阅