Violence detection using Oriented VIolent Flows

作者：

Highlights：

• A novel feature named OViF for violence detection.

• OViF makes full use of the orientation information of optical flows compared to ViF.

• We use AdaBoost as a feature selector and Linear SVM as a classifier.

• Experiments show that ViF + OViF obtain the state-of-the-art violence detection performance when using AdaBoost with SVM.

摘要

•A novel feature named OViF for violence detection.•OViF makes full use of the orientation information of optical flows compared to ViF.•We use AdaBoost as a feature selector and Linear SVM as a classifier.•Experiments show that ViF + OViF obtain the state-of-the-art violence detection performance when using AdaBoost with SVM.

论文关键词：Violence detection,Oriented VIolent Flows,AdaBoost,SVM

论文评审过程：Received 12 March 2015, Revised 13 January 2016, Accepted 14 January 2016, Available online 24 February 2016, Version of Record 25 March 2016.

论文官网地址：https://doi.org/10.1016/j.imavis.2016.01.006