Infrared and Laser Engineering, Volume. 51, Issue 6, 20210547(2022)
A video anomaly detection method based on deep autoencoding Gaussian mixture model
Figures & Tables(5)
Fig. 1. Flow chart of abnormal event detection method based on DAGMM
Table 1. Overview of benchmark datasets
View tableView in ArticleTable 1. Overview of benchmark datasets
Attributes UCSD Ped2 ShanghaiTech Frames 4560 317398 Scene Single Multi Labels Spatial & Temporal Spatial & Temporal Resolution 360×240 856×480 Anomalies Biker, cart, etc Chasing, brawling sudden motion, etc
Table 2. Comparison with the state of the art methods in terms of AUC%
View tableView in ArticleTable 2. Comparison with the state of the art methods in terms of AUC%
Method UCSD Ped2 ShanghaiTech MPPCA[ 3 ]69.3% - MDT[ 4 ]82.9% - MT-FRCN[ 5 ]92.2% - Conv2D-AE[ 10 ]85.0% 60.9% Conv3D-AE[ 10 ]91.2% - ConvLSTM-AE[ 20 ]88.1% - StackRNN[ 21 ]92.2% 68.0% Baseline[ 18 ]95.4% 72.8% Proposed method 95.7% 72.9%
Table 3. Influence of the number of Gaussian mixture components number
K on the experimental results of the UCSD Ped2 data set (frame-level AUC%)View tableView in ArticleTable 3. Influence of the number of Gaussian mixture components number
K on the experimental results of the UCSD Ped2 data set (frame-level AUC%)$ K $ AUC% 2 92.3% 4 94.5% 8 951% 16 95.7% 32 95.6% 64 95.7%
Tools
Get Citation
Copy Citation Text
Youkun Zhong, Haining Mo. A video anomaly detection method based on deep autoencoding Gaussian mixture model[J]. Infrared and Laser Engineering, 2022, 51(6): 20210547
Paper Information
Category: Photoelectric measurement
Received: Aug. 7, 2021
Accepted: --
Published Online: Dec. 20, 2022
The Author Email: Mo Haining (sunny_cj@21cn.com)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20