Graph-based adaptive and discriminative subspace learning for face image clustering

作者：

Highlights：

• We achieve the best results on the face dataset involving noise.

• Image alignment is introduced to the graph learning process.

• A weighting matrix is used to distinguish the roles of different features.

• The relationship between samples is used to learn the graph model.

摘要

•We achieve the best results on the face dataset involving noise.•Image alignment is introduced to the graph learning process.•A weighting matrix is used to distinguish the roles of different features.•The relationship between samples is used to learn the graph model.

论文关键词：Subspace clustering,Graph learning,Image alignment,Face recognition

论文评审过程：Received 23 November 2020, Revised 21 October 2021, Accepted 29 November 2021, Available online 11 December 2021, Version of Record 23 December 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.116359