Reverse erasure guided spatio-temporal autoencoder with compact feature representation for video anomaly detection

Conclusion In conventional video anomaly detection based on deep learning, the deep network is optimized without focus and the similarity between different normal frames is ignored. To alleviate these issues, we designed a dualencoder single-decoder network to reconstruct frames and proposed a train...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Science China. Information sciences Jg. 65; H. 9; S. 194101
Hauptverfasser: Zhong, Yuanhong, Chen, Xia, Jiang, Jinyang, Ren, Fan
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Beijing Science China Press 01.09.2022
Springer Nature B.V
Schlagworte:
Anomalies
Computer Science
Deep learning
Frames (data processing)
Information Systems and Communication Service
Methods
Moop
Task complexity
Training
ISSN:1674-733X, 1869-1919
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Conclusion In conventional video anomaly detection based on deep learning, the deep network is optimized without focus and the similarity between different normal frames is ignored. To alleviate these issues, we designed a dualencoder single-decoder network to reconstruct frames and proposed a training strategy involving reverse erasure based on the reconstruction error and deep SVDD to regularize the training of the network. With this training strategy, the proposed model achieved high performance in terms of both the AUC and EER. Future work will involve the application of our training strategy to more complex tasks.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1674-733X
1869-1919
DOI:10.1007/s11432-021-3444-9