3D-Convolutional Neural Network with Generative Adversarial Network and Autoencoder for Robust Anomaly Detection in Video Surveillance

As the surveillance devices proliferate, various machine learning approaches for video anomaly detection have been attempted. We propose a hybrid deep learning model composed of a video feature extractor trained by generative adversarial network with deficient anomaly data and an anomaly detector bo...

Full description

Saved in:
Bibliographic Details
Published in:International journal of neural systems Vol. 30; no. 6; p. 2050034
Main Authors: Shin, Wonsup, Bu, Seok-Jun, Cho, Sung-Bae
Format: Journal Article
Language:English
Published: Singapore 01.06.2020
Subjects:
3D CNN
generative adversarial network
Video anomaly detection
transfer learning
machine learning
ISSN:1793-6462, 1793-6462
Online Access:Get more information
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:As the surveillance devices proliferate, various machine learning approaches for video anomaly detection have been attempted. We propose a hybrid deep learning model composed of a video feature extractor trained by generative adversarial network with deficient anomaly data and an anomaly detector boosted by transferring the extractor. Experiments with UCSD pedestrian dataset show that it achieves 94.4% recall and 86.4% precision, which is the competitive performance in video anomaly detection.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:1793-6462
1793-6462
DOI:10.1142/S0129065720500343