CNN人脸关键点检测

CNN人脸关键点检测

参考文献《Deep Convolutional Network Cascade for Facial Point Detection》\

摘要

　提出一种利用精心设计的三级/三层(three-level)卷积网络估计面部关键点位置的新方法。

At each level，多网络(multiple networks)的输出是鲁棒性和精确估计的保证。

卷积网络的深层结构助于定位高精度的关键点，全局高层次的特征基于整个人脸面部在初始化阶段提取。

优势一，整个人脸的纹理上下文信息用来定位每个关键点。

优势二、因为网络被训练来用于同时预测所有的关键点，关键点间的几何约束被隐式的编码。

所以，该方法可以避免由歧义和数据损坏引起的局部极小(in difficult image samples due to occlusions, large pose variations, and extreme lightings）。The networks at the following two levels are trained to locally refine initial predictions and their inputs are limited to small regions around the initial predictions. Several network structures critical for accurate and robust facial point detection are investigated. Exten