The CR toolbox provides an easy, user-oriented, matlab package to compute the optical flow and feature tracking from video.

In particular, this toolbox contains the following features:

• KLT feature tracker (Tomasi-Kanade feature detector & Lucas-Kanade tracking)
• Optical flow from either KLT or image correlation (using the additional MatPIV toolbox)
• Bi-Orthogonal Decomposition of optical flow or tracked features
• A player to display all results
• A full set of user interfaces

###License

The CR-toolbox is released under the open-source CeCILL-B license. It is a BSD-like license under french law.

The license text is in the CR/Licence-en.txt file.

If you are publishing a scientific works that have use this toolbox, you should cite the following paper:

BARBACCI, A., DIENER, J., HÉMON, P., ADAM, B., DONÈS, N., REVERET, L., MOULIA, B. 2014 A robust videogrametric method for the velocimetry of wind-induced motion in trees. Agricultural and Forest Meteorology 184 :.220 - 229

###Documentation

See the CR-toolbox documentation website

To contact me, the prefered way is to use github issues. Otherwise you can send me an email.

###Project

This toolbox has been developped in a collaboration between

• LadHyX lab of Ecole Polytechnique
• Piaf team of INRA
• Evasion  team of INRIA

It was founded by the ANR research project Chêne-Roseau

The implementation has been done by Julien Diener.