摘要
目前,基于深度学习的视频异常检测方法都是在单一视角下对视频片段中的异常行为或异常事物进行检测,忽视了视角信息在视频异常检测中的重要性。在单一视角下,当异常事物被遮挡或异常行为不明显时,现有算法的性能将难以得到保证。为此,文中首次将视角转换的概念引入到视频异常检测中,通过级联网络结构在多视角下进行异常判断来提升模型的鲁棒性。针对受限于数据集没有多视角的监督信息,难以实现真正的显式的视角转换问题,提出了一种基于隐式视角转换的视频异常检测方法.对初步检测结果为正常的目标帧,利用其与特定帧的光流信息,通过光流映射实现目标帧到特定帧视角的隐式视角转换,并对视角转换后的目标帧进行二次异常检测。通过多个视角来判定目标帧是否异常,为视频异常检测提供了一种新的思路。实验结果表明,所提方法对异常数据的反应更灵敏,具有更鲁棒的正常数据拟合能力,在UCSD Ped2和CUHK Avenue数据集上的AUC值分别达到了97.0%和88.9%。
Existing deep learning-based video anomaly detection methods all detect anomalies in video clips under a single view,ignoring the importance of view information in video anomaly detection.Under a single view,when anomalies are occluded or not obvious,the performance of existing algorithms will suffer drops.To avoid this problem,the author firstly introduces the concept of view transformation into video anomaly detection,which improves the robustness of the model by judging abnormalities from multiple views.However,due to the lack of multi-view supervision information in the dataset,it is difficult to achieve explicit view transformation.Specifically,in order to reflect the idea of view transformation,the author proposes a video anomaly detection method based on implicit view transformation,using the optical flow information between frames to warp the implicit view information of the previous frame to the target frame,so as to realize the implicit view transformation from the target frame to the previous frame.And then,the method performs secondary anomaly detection on the target frame after view transformation.Experimental results show that the proposed method responds more sensitively to abnormal data and has a more robust normal data fitting ability.The AUC values on the UCSD Ped2 and CUHK Avenue datasets reached 97.0%and 88.9%,respectively.
作者
冷佳旭
谭明圮
胡波
高新波
LENG Jia-xu;TAN Ming-pi;HU Bo;GAO Xin-bo(Key Laboratory of Image Cognition,Chongqing University of Posts and Telecommunications,Chongqing 400065,China;Jiangsu Key Laboratory of Image and Video Understanding for Social Safety,Nanjing University of Science and Technology,Nanjing 210094,China;School of Optoelectronic Engineering,Chongqing University of Posts and Telecommunications,Chongqing 400065,China)
出处
《计算机科学》
CSCD
北大核心
2022年第2期142-148,共7页
Computer Science
基金
国家自然科学基金(62036007,62050175,62102057)
重庆市教委科学技术研究项目(KJQN-202100627)。
关键词
视频异常检测
隐式视角转换
光流映射
多视角检测
深度学习
Video anomaly detection
Implicit view transformation
Optical flow warp
Multi-view detection
Deep learning
