GitHub - biotrump/FaceForest: http://blog.gimiatlicho.webfactional.com/?page

Branches Tags

Name		Name	Last commit message	Last commit date
Latest commit History 3 Commits
data		data
.gitignore		.gitignore
CMakeLists.txt		CMakeLists.txt
Makefile		Makefile
README.txt		README.txt
demo.cpp		demo.cpp
eval_ffd.cpp		eval_ffd.cpp
face_forest.cpp		face_forest.cpp
face_forest.hpp		face_forest.hpp
face_utils.cpp		face_utils.cpp
face_utils.hpp		face_utils.hpp
feature_channel_factory.hpp		feature_channel_factory.hpp
forest.hpp		forest.hpp
head_pose_sample.cpp		head_pose_sample.cpp
head_pose_sample.hpp		head_pose_sample.hpp
image_sample.cpp		image_sample.cpp
image_sample.hpp		image_sample.hpp
mean_shift.hpp		mean_shift.hpp
multi_part_sample.cpp		multi_part_sample.cpp
multi_part_sample.hpp		multi_part_sample.hpp
opencv_serialization.hpp		opencv_serialization.hpp
split_gen.hpp		split_gen.hpp
thread_pool.hpp		thread_pool.hpp
timing.hpp		timing.hpp
train_ffd.cpp		train_ffd.cpp
train_headpose.cpp		train_headpose.cpp
tree.hpp		tree.hpp
tree_node.hpp		tree_node.hpp

Repository files navigation

// Author: Mathias Dantone, BIWI, ETH Zurich
// Email: mdantone@vision.ee.ethz.ch

// THE SOFTWARE COMES "AS IS", WITH NO WARRANTIES. THIS MEANS NO
// EXPRESS, IMPLIED OR STATUTORY WARRANTY, INCLUDING WITHOUT
// LIMITATION, WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A
// PARTICULAR PURPOSE, ANY WARRANTY AGAINST INTERFERENCE WITH YOUR
// ENJOYMENT OF THE SOFTWARE OR ANY WARRANTY OF TITLE OR
// NON-INFRINGEMENT. THERE IS NO WARRANTY THAT THIS SOFTWARE WILL
// FULFILL ANY OF YOUR PARTICULAR PURPOSES OR NEEDS. ALSO, YOU MUST
// PASS THIS DISCLAIMER ON WHENEVER YOU DISTRIBUTE THE SOFTWARE OR
// DERIVATIVE WORKS.

// When using this software, please acknowledge the effort that
// went into development by referencing the paper:
//
// Dantone M, Fanelli G, Gall J. and Van Gool L.,
// Real Time Facial Feature Detection using Conditional Regression Forest,
IEEE Conference on Computer Vision and Pattern Recognition (CVPR'12), 2012.

// Note that this is not the original software that was used for
// the paper mentioned above. It is a re-implementation with different features.

This package contains the source code for training and evaluation of the
Conditional Regression Forest.

Additional to the source code you can find pretrained trees for head pose estimation
and also for facial feature detection.

___ BUILDING
This framework needs the open source computer vision library OpenCV and boost.

1) set the path to OpenCV and Boost the Makefile
2) make all (compiling the source code)
3) make clear

Tested with the following version of the third party lib:
- OpenCV 2.3.X
- Boost 1.47.0 + Boost 1.46.1

Known problems:
- The boost library can be build in different ways.
Sometimes the boost shared libs have the ending '-mt' and sometimes they don't.
Please fix the makefile according to your version of boost.

Tested on the following operation system:
- MacOS 10.7.4
- ubuntu 12.04 LTS
- Debian 2.30.2

___ DEMO APPLICATION
Running the demo application using the pretrained trees is easy.
Just run the application 'FaceForest':
./FaceForest 0 data/config_ffd.txt data/config_headpose.txt data/haarcascade_frontalface_alt.xml

you need to set 4 flags:
- mode (0==training, 1==evaluate)
- path to ffd config file
- path to headpose config file
- path to face cascade