/NIID-MSL

Robust Multi-view Subspace Learning with Non-independent and Non-identically Distributed Complex Noise (TNNLS, 2019)

Primary LanguageMATLAB

NIID-MSL

This is the source code of the paper "Robust Multi-view Subspace Learning with Non-independent and Non-identically Distributed Complex Noise". We provided two implementation of the proposed NIID-MSL in the floder ./varaiatoin, hdp_multi_view.m and hdp_multi_view_cell.m. The former one is followed the regular VB only using MATLAB and latter one is accelerated through the strategies of the Section V.C and mix-programing with Matlab and C. If you have any questions, please feel free to contact me (zsy20110806207@stu.xjtu.edu.cn).

Experimental Environment

  • MATLAB 2018a
  • Ubuntu 16.04
  • C Compiler: Gcc 6.3

How to run

  1. Download the dataset from the following link and put them in the ./data floder.
  1. Switch to the root path of thei project and add path to matlab
 init_path 
  1. Run any of the demo listed in the ./demo floder, such as
 demo_scurve