/Adaptive-Low-Rank-Tensor-Representation

Matlab implementation of TNNLS2019 paper " Accurate Tensor Completion via Adaptive Low-Rank Representation "

Primary LanguageC++

Adaptive-Low-Rank-Tensor-Representation

Matlab implementation of TNNLS2019 paper: "Accurate Tensor Completion via Adaptive Low-Rank Representation"

Dependencies

  • Matlab 2017
  • tensor_toolbox 2.6

Preparation

  1. We provide the library tensor_toolbox_2.6, please decompress the file 'tensor_toolbox_2.6.tar.gz' using the following command line
cd ./Adaptive-Low-Rank-Tensor-Representation/
tar -zxvf tensor_toolbox_2.6.tar.gz
  1. Run the matlab file './script/CompileFile.m' to mex compile the CPP file in the sampling method. We have provided the library files compiled on 64 bit Ubuntu OS, which can be found in the folder './script/'.

Usage

Run the matlab file 'main_Inpainting.m' to see the demo on image inpainting.

Reference

If you find our work useful in your research or publication, please cite our work:
[1] Lei Zhang, Wei Wei, Qinfeng Shi, Chunhua Shen, Anton van den Hengel, and Yanning Zhang. "Accurate Tensor Completion via Adaptive Low-Rank Representation." IEEE Transactions on Neural Networks and Learning Systems (2019).[PDF]

@article{zhang2019accurate,
  title={Accurate Tensor Completion via Adaptive Low-Rank Representation},
  author={Zhang, Lei and Wei, Wei and Shi, Qinfeng and Shen, Chunhua and van den Hengel, Anton and Zhang, Yanning},
  journal={IEEE Transactions on Neural Networks and Learning Systems},
  year={2019},
  publisher={IEEE}
}