Video anomaly detection and localization using motion-field shape description and homogeneity testing

作者：

Highlights：

• We introduce a histogram-based shape descriptor to motion field in each local patch.

• The motion descriptor captures the motion trend and details in local patches.

• We propose a similarity-based statistical model to detect spatio-temporal anomalies.

• The statistical model relies on unsupervised learning without any prior assumption.

• The method can adapt to the whole scene with tolerance to perspective distortion.

摘要

•We introduce a histogram-based shape descriptor to motion field in each local patch.•The motion descriptor captures the motion trend and details in local patches.•We propose a similarity-based statistical model to detect spatio-temporal anomalies.•The statistical model relies on unsupervised learning without any prior assumption.•The method can adapt to the whole scene with tolerance to perspective distortion.

论文关键词：Abnormal activity,Anomaly detection,Anomaly localization,Shape description,K-NN similarity-based outlier detection

论文评审过程：Received 22 August 2018, Revised 19 March 2020, Accepted 21 April 2020, Available online 27 April 2020, Version of Record 8 May 2020.

论文官网地址：https://doi.org/10.1016/j.patcog.2020.107394