Anomaly Detection in Moving Crowds through Spatiotemporal Autoencoding and Additional Attention
We propose an anomaly detection approach by learning a generative model using deep neural network. A weighted convolutional autoencoder- (AE-) long short-term memory (LSTM) network is proposed to reconstruct raw data and perform anomaly detection based on reconstruction errors to resolve the existin...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2018-01-01
|
| Series: | Advances in Multimedia |
| Online Access: | http://dx.doi.org/10.1155/2018/2087574 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832551270883262464 |
|---|---|
| author | Biao Yang Jinmeng Cao Rongrong Ni Ling Zou |
| author_facet | Biao Yang Jinmeng Cao Rongrong Ni Ling Zou |
| author_sort | Biao Yang |
| collection | DOAJ |
| description | We propose an anomaly detection approach by learning a generative model using deep neural network. A weighted convolutional autoencoder- (AE-) long short-term memory (LSTM) network is proposed to reconstruct raw data and perform anomaly detection based on reconstruction errors to resolve the existing challenges of anomaly detection in complicated definitions and background influence. Convolutional AEs and LSTMs are used to encode spatial and temporal variations of input frames, respectively. A weighted Euclidean loss is proposed to enable the network to concentrate on moving foregrounds, thus restraining background influence. Moving foregrounds are segmented from the input frames using robust principal component analysis decomposition. Comparisons with state-of-the-art approaches indicate the superiority of our approach in anomaly detection. Generalization of anomaly detection is improved by enforcing the network to focus on moving foregrounds. |
| format | Article |
| id | doaj-art-a733d372910246c1a3094adf2af5ebe0 |
| institution | Kabale University |
| issn | 1687-5680 1687-5699 |
| language | English |
| publishDate | 2018-01-01 |
| publisher | Wiley |
| record_format | Article |
| series | Advances in Multimedia |
| spelling | doaj-art-a733d372910246c1a3094adf2af5ebe02025-02-03T06:01:50ZengWileyAdvances in Multimedia1687-56801687-56992018-01-01201810.1155/2018/20875742087574Anomaly Detection in Moving Crowds through Spatiotemporal Autoencoding and Additional AttentionBiao Yang0Jinmeng Cao1Rongrong Ni2Ling Zou3Department of Information Science and Engineering, Changzhou University, Changzhou, Jiangsu, ChinaDepartment of Information Science and Engineering, Changzhou University, Changzhou, Jiangsu, ChinaChangzhou Vocational Institute Textile and Garment, Changzhou, Jiangsu, ChinaDepartment of Information Science and Engineering, Changzhou University, Changzhou, Jiangsu, ChinaWe propose an anomaly detection approach by learning a generative model using deep neural network. A weighted convolutional autoencoder- (AE-) long short-term memory (LSTM) network is proposed to reconstruct raw data and perform anomaly detection based on reconstruction errors to resolve the existing challenges of anomaly detection in complicated definitions and background influence. Convolutional AEs and LSTMs are used to encode spatial and temporal variations of input frames, respectively. A weighted Euclidean loss is proposed to enable the network to concentrate on moving foregrounds, thus restraining background influence. Moving foregrounds are segmented from the input frames using robust principal component analysis decomposition. Comparisons with state-of-the-art approaches indicate the superiority of our approach in anomaly detection. Generalization of anomaly detection is improved by enforcing the network to focus on moving foregrounds.http://dx.doi.org/10.1155/2018/2087574 |
| spellingShingle | Biao Yang Jinmeng Cao Rongrong Ni Ling Zou Anomaly Detection in Moving Crowds through Spatiotemporal Autoencoding and Additional Attention Advances in Multimedia |
| title | Anomaly Detection in Moving Crowds through Spatiotemporal Autoencoding and Additional Attention |
| title_full | Anomaly Detection in Moving Crowds through Spatiotemporal Autoencoding and Additional Attention |
| title_fullStr | Anomaly Detection in Moving Crowds through Spatiotemporal Autoencoding and Additional Attention |
| title_full_unstemmed | Anomaly Detection in Moving Crowds through Spatiotemporal Autoencoding and Additional Attention |
| title_short | Anomaly Detection in Moving Crowds through Spatiotemporal Autoencoding and Additional Attention |
| title_sort | anomaly detection in moving crowds through spatiotemporal autoencoding and additional attention |
| url | http://dx.doi.org/10.1155/2018/2087574 |
| work_keys_str_mv | AT biaoyang anomalydetectioninmovingcrowdsthroughspatiotemporalautoencodingandadditionalattention AT jinmengcao anomalydetectioninmovingcrowdsthroughspatiotemporalautoencodingandadditionalattention AT rongrongni anomalydetectioninmovingcrowdsthroughspatiotemporalautoencodingandadditionalattention AT lingzou anomalydetectioninmovingcrowdsthroughspatiotemporalautoencodingandadditionalattention |