Supervised Descent Method Apply to Face Alignment, and Head Pose Estimation with Linear Regression. It is cross-platfrom, easily compile in windows, ubuntu, even in Android & iOS.
git clone https://github.com/chengzhengxin/sdm.git
cd sdm/example/ ./Roboman-SDM-TEST.exe
You will look like as follows:
Make sure you have installed the cmake-gui & Qt5.3.2 (mingw will be needed).
mkdir build cd build cmake-gui mingw32-make
Maybe you should modify the CMakeLists.txt about OpenCV libs
mkdir build cd build cmake .. make cp Roboman-SDM-TEST ../example/ cd ../example/ ./Roboman-SDM-TEST
1.Copy "haarrobomanff_alt2.xml" & "roboman-landmark-model.bin" to your iOS Project.
2.Add the OpenCV2.framework to you iOS Project, make sure you can easily read camera and show frame with OpenCV.
3.Copy the include folder under src folder to you iOS Project source code, just copy, not anything else.
4.And then "#include "include/ldmarkmodel.h"" in your .mm file, make sure not in .h file, use it like test_model.cpp, Compile & enjoy it.
5.Make sure iOS project Build Settings option "Enable Bitcode" is "NO" (above XCode7).
According to Android-cmake-Tool.