Source Code

GIST3d

The provided codes are meant for research purposes only. Generated output is a holistic video descriptor for any given video clip or a spatio-temporal volume, representing the motion and scene components.

The example codes for descriptor computation and visualization routines are provided as-it-is without any warranty. For questions and bug reports please contact the authors at bsolmaz@cs.ucf.edu. The password is: ucf

Matlab Codes

C++ Codes

For more details, please see the enclosed readme file within the provided codes and/check the project page at http://crcv.ucf.edu/projects/gist3d/

Related Publication:
Berkan Solmaz, Shayan M. Assari, and Mubarak Shah, Classifying Web Videos using a Global Video Descriptor, Machine Vision and Applications (MVA), 24 (7), 1473-1485, 2012 Descriptor, Machine vision and applications 24 (7), 1473-1485, 2012.