Multi-view facial landmark detection by using a 3D shape model

作者：

Highlights：

• Accurate facial landmark detector coupled with a 6 DoF head pose estimator

• A reliable global method followed by a precise local optimization

• Novel learning method based on structured output SVM

• Multi-view detector working on a wide range of viewing angles (frontal-profile). Self-occlusions handled.

• Comparison with recent state-of-the-art methods on standard “in the wild” datasets with quantitatively favourable results.

摘要

•Accurate facial landmark detector coupled with a 6 DoF head pose estimator•A reliable global method followed by a precise local optimization•Novel learning method based on structured output SVM•Multi-view detector working on a wide range of viewing angles (frontal-profile). Self-occlusions handled.•Comparison with recent state-of-the-art methods on standard “in the wild” datasets with quantitatively favourable results.

论文关键词：Face,Landmarks,Detection,Localization,3D model,Shape,Occlusions

论文评审过程：Received 21 December 2014, Revised 16 September 2015, Accepted 30 November 2015, Available online 15 December 2015, Version of Record 17 May 2016.

论文官网地址：https://doi.org/10.1016/j.imavis.2015.11.003