Skip to content
Multi-view Hybrid Embedding: A Divide-and-Conquer Approach
MATLAB
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
GetZeroMeanOneVar.p
MvPCA.m
MvPCA_Projection.m
PCA.p
ZeroMeanOneVar.p
construct_L.m
demo.asv
demo.m
nearest_neighbor_index.m
pie.mat
preprocessing.m
proposed_method.asv
proposed_method.m
readme.txt
testing.m

readme.txt

Data
pie.mat
Tr_data <1x5 cell>    % training data
Tr_Labels <1x5 cell>  % labels of training data
Te_data <1x5 cell>    % test data
Te_Labels <1x5 cell>  % labels of test data


Dependencies
Matlab 2013a % please replace knnclassify function in testing.m file and knnsearch function in nearest_neighbor_index.m file when another version is used.


Demo
"demo.m" is a demo for the training and testing on CMU PIE dataset. 


Citation
@article{xu2019multiview,
  title={Multiview Hybrid Embedding: A Divide-and-Conquer Approach},
  author={Xu, Jiamiao and Yu, Shujian and You, Xinge and Leng, Mengjun and Jing, Xiao-Yuan and Chen, CL Philip},
  journal={IEEE transactions on cybernetics},
  year={2019},
  publisher={IEEE}
}



Contact
If any problem or suggestion, please contact Jiamiao Xu (JiamiaoXu_93@163.com)


The source codes and processed data can only be used for non-commercial purpose. Please make appropriate reference to our work if you are using the codes for your research. Thank you.
You can’t perform that action at this time.