Robust salient motion detection in non-stationary videos via novel integrated strategies of spatio-temporal coherency clues and low-rank analysis

作者：

Highlights：

• We propose a versatile and robust background tracking method.

• We integrate low-rank background prior with coherency revealing based on the aligned RPCA low-rank analysis.

• We define a series of saliency clues via online exploration of the spatio-temporal coherency.

摘要

Highlights•We propose a versatile and robust background tracking method.•We integrate low-rank background prior with coherency revealing based on the aligned RPCA low-rank analysis.•We define a series of saliency clues via online exploration of the spatio-temporal coherency.

论文关键词：Salient motion detection,Non-stationary videos,Aligned RPCA,Low-rank analysis,Semantic coherency,Monitor stabilization

论文评审过程：Received 13 May 2015, Revised 18 September 2015, Accepted 27 September 2015, Available online 22 October 2015, Version of Record 24 December 2015.

论文官网地址：https://doi.org/10.1016/j.patcog.2015.09.033