cjmcv

论文：

1. Ren S, Cao X, Wei Y, et al. Face Alignment at 3000 FPS via Regressing Local Binary Features[C]// IEEE Conference on Computer Vision and Pattern Recognition. IEEE, 2014:1685-1692.

2. Chen D, Ren S, Wei Y, et al. Joint Cascade Face Detection and Alignment[M]// Computer Vision – ECCV 2014. Springer International Publishing, 2014:109-122.

工程代码：

1. MATLAB: https://github.com/jwyang/face-alignment

2. C++: https://github.com/yulequan/face-alignment-in-3000fps

博客推荐:

• http://www.cnblogs.com/nsnow/p/4590509.html

算法使用感觉

速度较快，特征点跳动较小。使用从一个场景中收集的训练样本进行训练后，得到的模型能够较好地应用在该场景中，但场景更换后效果会下降，对场景依赖较大。