SeetaFace中科院开源人脸项目Ubuntu下编译

SeetaFace Engine is an open source C++ face recognition engine,which can run on cpu with no third-party dependence. It contains three key parts,i.e.,SeetaFace Detection,SeetaFace Alignment and SeetaFace Identification,which are necessary and sufficient for building a real-world face recognition applicaiton system.

1.下载

git clone https://github.com/seetaface/SeetaFaceEngine.git

2.安装cmake工具

CMake 3.1.0 or higher is required

ubunut14.04自带的cmake版本太低，可以去官网自行下载安装

3.编译FaceDetection

进入FaceDetection路径下

mkdir linux

进入linux文件夹下面

cmake ..

make

生成可执行文件facedet_test

4.编译FaceIdentification

进入FaceIdentification路径下

拷贝FaceDetection include下的face_detection.h 到include下面

mkdir linux

修改CMakeLists.txt

同时将FaceDetection 生成的libseeta_facedet_lib.so 放在新建的linux文件夹下面
进入到linux文件夹
cmake ..
make
生成fa_test

执行效果如下：

5.编译FaceIdentification
进入FaceIdentification路径下
（1）修改src/test/CMakeLists.txt
添加前两个项目生成的so

12   add_executable(${BIN} ${f})
 13   target_link_libraries(${BIN} viplnet ${OpenCV_LIBS})
 14         target_link_libraries(${BIN} viplnet ${OpenCV_LIBS} /root/1xiu/face/SeetaFaceEngine-master/FaceIdentification/linux/libseeta_facedet_lib.so)
 15         target_link_libraries(${BIN} viplnet ${OpenCV_LIBS} /root/1xiu/face/SeetaFaceEngine-master/FaceIdentification/linux/libseeta_fa_lib.so)

（2）拷贝FaceDetection include下的face_detection.h 和FaceIdentification include下面的face_alignment.h到FaceIdentification的include下面

（3）新建文件夹linux
进入到linux
cmake ..
make
生成src/test/test_face_verification.bin
在实际执行测试的时候需要修改*test.cpp，

将比对的文件修改为命令行输入

执行结果如下：

./src/test/test_face_verification.bin  ../../FaceDetection/pic2.jpg ../../FaceDetection/pic3.jpg 
0.700385