Design and learn distinctive features from pore-scale facial keypoints

作者：

Highlights：

• A new framework for pore-scale feature-extraction and description is proposed.

• A pore-to-pore correspondences dataset is established.

• A learning-based pore-scale facial-feature method is proposed.

• The proposed method can be applied to face recognition, 3D reconstruction, etc.

摘要

Highlights•A new framework for pore-scale feature-extraction and description is proposed.•A pore-to-pore correspondences dataset is established.•A learning-based pore-scale facial-feature method is proposed.•The proposed method can be applied to face recognition, 3D reconstruction, etc.

论文关键词：Pore-scale facial feature,Feature extraction,Face matching,Face reconstruction,Face recognition

论文评审过程：Received 31 May 2014, Revised 26 August 2014, Accepted 27 September 2014, Available online 16 October 2014.

论文官网地址：https://doi.org/10.1016/j.patcog.2014.09.026