Code for paper "Constructing Robust Affinity Graphs for Spectral Clustering", adapted from MMLAB, CUHK
This is the demo of the following paper on how to construct affinity graphs.
The random forest inlcuded in this code is a revised version based on the randomforest-matlab project:
http://code.google.com/p/randomforest-matlab/
The code of spectral clustering algorithm is provided from:
http://www.vision.caltech.edu/lihi/Demos/SelfTuning/ZPclustering.zip
Reference:
[1] Constructing Robust Affinity Graphs for Spectral Clustering
X. Zhu, C.C. Loy and S. Gong.
In Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Columbus, Ohio, USA, June 2014.