人脸对齐：人脸关键点检测发展概述（含数据集及论文解读）

4个月的实习结束了，本文主要记录自己实习期间对人脸关键点检测的调研结果。

目录结构

1.背景介绍

2.开源数据集

3.评价指标

4.检测方法

5.论文解读

5.1传统方法

5.1.1 Active Shape Models-Their Training and Application

5.2基于级联形状回归

5.2.1 Cascaded pose regression CVPR2010

5.3 基于深度学习

5.3.1 Deep Convolutional Network Cascade for Facial Point Detection CVPR2013

5.3.2 Extensive Facial Landmark Localization with Coarse-to-Fine Convolutional Network Cascade ICCV2013

5.3.3 Joint Face Detection and Alignment using Multi-task Cascaded Convolutional Networks ECCV2016

5.3.4 Deep Alignment Network: A convolutional neural network for robust face alignment CVPR2017

5.4 前沿方法

5.4.1 Look at Boundary: A Boundary-Aware Face Alignment Algorithm CVPR2018

5.4.2 Style Aggregated Network for Facial Landmark Detection CVPR2018

5.4.3 Semantic Alignment: Finding Semantically Consistent Ground-Truth for Facial Landmark Detection CVPR2019

5.4.4 Robust Facial Landmark Detection via Occlusion-adaptive Deep Networks CVPR2019

6.部分论文复现成果展示

公开数据集



真实监控场景