Multiple structure recovery via robust preference analysis

作者：

Highlights：

• A method to extract multiple structure from outlier contaminated data is presented.

• Preference analysis, matrix decomposition and robust statistic are combined together.

• The method proven successful in many real word applications.

摘要

•A method to extract multiple structure from outlier contaminated data is presented.•Preference analysis, matrix decomposition and robust statistic are combined together.•The method proven successful in many real word applications.

论文关键词：Multi-model fitting,Model estimation,Spectral clustering,Matrix factorization

论文评审过程：Received 13 July 2016, Revised 24 May 2017, Accepted 9 September 2017, Available online 20 September 2017, Version of Record 25 September 2017.

论文官网地址：https://doi.org/10.1016/j.imavis.2017.09.005