Facial landmarking for in-the-wild images with local inference based on global appearance

作者：

Highlights：

• The problem of facial landmark detection in unconstrained conditions is tackled.

• A new way of constructing per-landmark response maps is proposed.

• The method is based on the selection at test time of globally similar images.

• We use local Gaussian process regression for inference.

摘要

•The problem of facial landmark detection in unconstrained conditions is tackled.•A new way of constructing per-landmark response maps is proposed.•The method is based on the selection at test time of globally similar images.•We use local Gaussian process regression for inference.

论文关键词：Facial landmarking,Regression,Part-based models,Gaussian processes

论文评审过程：Received 27 February 2014, Revised 25 November 2014, Accepted 6 January 2015, Available online 21 February 2015.

论文官网地址：https://doi.org/10.1016/j.imavis.2015.01.004