FaceTracker

FaceTracker is a library for deformable face tracking written in C++ using OpenCV 2, authored by Jason Saragih and maintained by Kyle McDonald.

It is available free for non-commercial use, and may be redistributed under these conditions. Please see license.md for complete details. For commercial use, request a quote.

Wrappers are available for:

• openFrameworks: ofxFaceTracker
• Cinder: ciFaceTracker
• Python: pyfacetracker

Installation

These instructions are for compiling the code on OS X only. Compilation on other Unix-type architectures should be similar.

1. Download the latest version of OpenCV-2.0 with git clone git://code.opencv.org/opencv.git (more instructions here)
2. Follow the OpenCv installation instructions and compile and install OpenCV to some arbitrary local directory (usually /usr/local/).
3. Clone the FaceTracker code with git clone git://github.com/kylemcdonald/FaceTracker.git. This will number of subdirectories within the root directory:
   • src (contains all source code)
   • model (contains a pre-trained tracking model)
   • bin (will contain the executable after building)
4. In the Makefile located in the root directory of the FaceTracker code, change the OPECV_PATH variable to the directory where OpenCV was installed (by defualt /usr/local). Optionally, you can also add -fopenmp to the CFLAGS and -lgomp to LIBRARIES to compile with OpenMP support.
5. Build the system with make.
6. The executable face_tracker can be found in the bin subdirectory.

To build on Ubuntu Linux you need:

sudo apt-get install libcv-dev libopencv-dev

face_tracker Usage

Usage: face_tracker [options]
Options:
-m <string> : Tracker model (default: ../model/face2.tracker)
-c <string> : Connectivity (default: ../model/face.con)
-t <string> : Triangulation (default: ../model/face.tri)
-s <double> : Image scaling (default: 1)
-d <int>    : Frames/detections (default: -1)
--check     : Check for failure 
--help      : Print help
-?          : Print help