期刊
IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY
卷 14, 期 5, 页码 1390-1399出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TIFS.2018.2878538
关键词
Spatio-temporal; anomaly detection; variational autoencoder; loss function
资金
- National Natural Science Foundation of China [61503017, U1435220, 61866022, 61802180]
- Aeronautical Science Foundation of China [2016ZC51022]
- SURECAP CPER Project
- EU Horizon 2020 Research and Innovation Programme [690238]
- UK EPSRC [EP/P031668/1]
- BT Ireland Innovation Centre (BTIIC)
- Platform CAPSEC - Region ChampagneArdenne
- FEDER
- National Research Foundation of Korea (NRF) - Korea Government (Ministry of Science and ICT) [2017R1E1A1A01077913]
- H2020 Societal Challenges Programme [690238] Funding Source: H2020 Societal Challenges Programme
Security surveillance is critical to social harmony and people's peaceful life. It has a great impact on strengthening social stability and life safeguarding. Detecting anomaly timely, effectively and efficiently in video surveillance remains challenging. This paper proposes a new approach, called S-2-VAE, for anomaly detection from video data. The S-2-VAE consists of two proposed neural networks: a Stacked Fully Connected Variational AutoEncoder (S-F-VAE) and a Skip Convolutional VAE (S-C-VAE). The S-F-VAE is a shallow generative network to obtain a model like Gaussian mixture to fit the distribution of the actual data. The S-C-VAE, as a key component of S(2-)VAE, is a deep generative network to take advantages of CNN, VAE and skip connections. Both S-F-VAE and S-C-VAE are efficient and effective generative networks and they can achieve better performance for detecting both local abnormal events and global abnormal events. The proposed S-2-VAE is evaluated using four public datasets. The experimental results show that the S-2-VAE outperforms the state-of-the-art algorithms. The code is available publicly at https://github.com/tianwangbuaa/.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据