Constrained mutual convex cone method for image set based recognition

作者：

Highlights：

• Propose novel image-set classification frameworks based on convex cone representation.

• Define the similarity between two convex cones.

• Introduce discriminative feature extraction method based on the gaps among convex cones.

• Demonstrate the effectiveness of the proposed methods through visualizations and classification experiments.

摘要

•Propose novel image-set classification frameworks based on convex cone representation.•Define the similarity between two convex cones.•Introduce discriminative feature extraction method based on the gaps among convex cones.•Demonstrate the effectiveness of the proposed methods through visualizations and classification experiments.

论文关键词：Image-set based method,Convex cone representation,Multiple angles

论文评审过程：Received 25 September 2020, Revised 17 June 2021, Accepted 4 July 2021, Available online 28 July 2021, Version of Record 15 August 2021.

论文官网地址：https://doi.org/10.1016/j.patcog.2021.108190