This project is forked from FasterRCNN-Encapsulation-Cplusplus .
Please refer to below blogs for detail explanations.
faster_rcnn c++版本的 caffe 封装（1）
http://www.cnblogs.com/louyihang-loves-baiyan/p/5485955.html
faster_rcnn c++版本的 caffe 封装，动态库（2）
http://www.cnblogs.com/louyihang-loves-baiyan/p/5493344.html
将Faster RCNN的python demo改成C++ demo
http://blog.csdn.net/xyy19920105/article/details/50440957
I added some local modifications to make it run on my Ubuntu 14.04 system, mostly on directory faster_cxx_lib.
Original project use OpenCV3, I downgrade to use OpenCV2 (I remove all libopencv_imgcodecs dependencies in make scripts)
To use this demo, you will need to put directory faster_cxx_lib under py-faster-rcnn/ .
cd to faster_cxx_lib, first run ./make_all.sh, then run ./run_main.sh
It can run on both py-faster-rcnn's alt-opt and end2end caffemodels.