SVAS: Surveillance Video Analysis System

作者：

Highlights：

• A Surveillance Video Analysis System (SVAS) is proposed.

• An Interval-Based Spatio-Temporal Model (IBSTM) is introduced.

• Automatic learning, detection and inference of complex video events are provided.

• Hybrid machine learning methods are used and extended via Threshold Models.

• The empirical evaluation shows that SVAS succeeds in complex event recognition.

摘要

•A Surveillance Video Analysis System (SVAS) is proposed.•An Interval-Based Spatio-Temporal Model (IBSTM) is introduced.•Automatic learning, detection and inference of complex video events are provided.•Hybrid machine learning methods are used and extended via Threshold Models.•The empirical evaluation shows that SVAS succeeds in complex event recognition.

论文关键词：Event detection,Markov logic networks,Video surveillance,Event model learning,Event inference

论文评审过程：Received 19 January 2017, Revised 28 July 2017, Accepted 29 July 2017, Available online 29 July 2017, Version of Record 4 August 2017.

论文官网地址：https://doi.org/10.1016/j.eswa.2017.07.051