Multi-property tensor-based learning for abnormal event detection

Λεπτομέρειες βιβλιογραφικής εγγραφής
Τίτλος: Multi-property tensor-based learning for abnormal event detection
Συγγραφείς: Bakalos, Nikolaos, Doulamis, Nikolaos, Doulamis, Anastasios, Makantasis, Konstantinos, International Symposium on Visual Computing ISVC 2022
Στοιχεία εκδότη: Springer International Publishing
Έτος έκδοσης: 2022
Συλλογή: University of Malta: OAR@UM / L-Università ta' Malta
Θεματικοί όροι: Video surveillance -- Data processing, Event processing (Computer science), Image processing -- Data processing, Tensor algebra
Περιγραφή: In this paper, we propose a novel abnormal event detection scheme for video surveillance systems using an unsupervised learning process. Our contribution includes intra and inter property feature encoding to reduce information redundancy within (intra) and across (inter) image features. Intra property encoding is carried out using convolutional auto-encoders. Inter-property encoding is performed using an unsupervised tensor-based learning mode to handle the dimensionality issue arising in cases when different properties are inter-related together. Comprehensive experiments are performed on two benchmarks:Avenue, and ShanghaiTech. ; peer-reviewed
Τύπος εγγράφου: conference object
Γλώσσα: English
Relation: https://www.um.edu.mt/library/oar/handle/123456789/125534
DOI: 10.1007/978-3-031-20713-6_25
Διαθεσιμότητα: https://www.um.edu.mt/library/oar/handle/123456789/125534
https://doi.org/10.1007/978-3-031-20713-6_25
Rights: info:eu-repo/semantics/restrictedAccess ; The copyright of this work belongs to the author(s)/publisher. The rights of this work are as defined by the appropriate Copyright Legislation or as modified by any successive legislation. Users may access this work and can make use of the information contained in accordance with the Copyright Legislation provided that the author must be properly acknowledged. Further distribution or reproduction in any format is prohibited without the prior permission of the copyright holder.
Αριθμός Καταχώρησης: edsbas.5E36BC33
Βάση Δεδομένων: BASE