We are excited to announce a new facial behaviour analysis tool OpenFace! It is more accurate, more stable and better documented than CLM-framework and can be seen as the next step of the project. You can find the code for our new OpenFace tool here - https://github.com/TadasBaltrusaitis/OpenFace
All the continuing developement will happen on OpenFace, but for backwards compatibility we are keeping CLM-framework available on github.
Framework for various Constrained Local Model based face tracking and landmark detection algorithms and their extensions/applications. Includes CLM, CLM-Z and CLNF algorithms. More details can be found in Readme.txt.
The framework also includes a brand new Facial Action Unit detection in videos system (see Readme_action_units.txt).
The framework also includes a brand new gaze estimation system as well (see Readme_gaze.txt).
The code was written mainly by Tadas Baltrusaitis during his time at the Language Technologies Institute at the Carnegie Mellon University; Computer Laboratory, University of Cambridge; and Institute for Creative Technologies, University of Southern California.
Special thanks goes to Louis-Philippe Morency and his MultiComp Lab at Institute for Creative Technologies for help in writing and testing the code, and Erroll Wood for the gaze estimation work.
The stable versions of the framework have been tagged, the latest version of Cambridge Face Tracker is is 1.3.6 and I recommend you download it.
Some examples of the system in action: http://youtu.be/V7rV0uy7heQ http://youtu.be/vYOa8Pif5lY http://youtu.be/LDBu0BLKVDw
For Windows systems open and compile CLM_framework_vs2013.sln (requires VisualStudio 2012) or CLM_framework_vs2013.sln (requires VisualStudio 2013). All the required libraries are included with the code.
For Unix based systems follow readme-ubuntu.txt
For Windows systems you can find the compiled binaries here: http://www.cl.cam.ac.uk/~tb346/software/Cambridge_Face_Tracker_1.3.6.zip