panji530 / EDSC

Efficient dense subspace clustering

EDSC

Code for the following paper:
Ji, P., Salzmann, M., Li, H.: Efficient dense subspace clustering. In: IEEE WACV, PP 461-468 (2014).

1. This code uses the normalized cuts algorithm for the final clustering. Please download it from http://www.cis.upenn.edu/~jshi/software/ https://github.com/panji1990/Ncut9, compile it properly and add its path in your Matlab.

2. To test the code on Hopkins155 motion segmentation dataset, please download the dataset from http://vision.jhu.edu/data/ and then run the file run_Hopkins155.m .

3. To test the code on Extended YaleB, please run the file run_YaleFaces.m .

4. If you have any question regarding the code, please contact Pan Ji at [e-mail] .