Real-time nonparametric background subtraction with tracking-based foreground update

作者：

Highlights：

• Combined foreground-background spatio-temporal nonparametric models with tracking-based update of reference data.

• Bayesian classifier for combining foreground and background models with different spatial bandwidths.

• Selective analysis strategy based on random sampling and regions of interest.

• Efficient automatic appearance bandwidth selection switching.

• Real-time, GPU-based implementation of the proposed strategy.

摘要

•Combined foreground-background spatio-temporal nonparametric models with tracking-based update of reference data.•Bayesian classifier for combining foreground and background models with different spatial bandwidths.•Selective analysis strategy based on random sampling and regions of interest.•Efficient automatic appearance bandwidth selection switching.•Real-time, GPU-based implementation of the proposed strategy.

论文关键词：Foreground segmentation,Background subtraction,Nonparametric modelling,Parallel processing,Real-time GPU

论文评审过程：Received 22 December 2016, Revised 17 July 2017, Accepted 5 September 2017, Available online 14 September 2017, Version of Record 22 September 2017.

论文官网地址：https://doi.org/10.1016/j.patcog.2017.09.009