Classification and saliency detection by semi-supervised low-rank representation

作者：

Highlights：

• A semi-supervised version of low rank representation is presented.

• Label information based outlier and saliency detection strategies are presented.

• Semi-supervised low rank learning is efficient and effective in classification.

• Semi-supervised low rank learning is reliable in outlier detection.

摘要

Highlights•A semi-supervised version of low rank representation is presented.•Label information based outlier and saliency detection strategies are presented.•Semi-supervised low rank learning is efficient and effective in classification.•Semi-supervised low rank learning is reliable in outlier detection.

论文关键词：Low rank representation,Semi-supervised learning,Outlier detection,Saliency detection

论文评审过程：Received 10 June 2014, Revised 24 July 2015, Accepted 6 September 2015, Available online 28 September 2015, Version of Record 27 November 2015.

论文官网地址：https://doi.org/10.1016/j.patcog.2015.09.008