Low-Rank Tensor Recovery via Variational Schatten-$p$ Quasi-Norm and Jacobian Regularization

Installation

1. Download source code and dataset:

   • git clone https://github.com/CZY-Code/CP-Pruner.git
   • Download the datasets
     • ShapeNet
     • Color Image
     • CAVE
     • WDC Mall and PaviaU
     • Videos
     • Bunny

2. Pip install dependencies:

   • OS: Ubuntu 20.04.6
   • nvidia :
     • cuda: 12.1
     • cudnn: 8.5.0
   • python == 3.9.18
   • pytorch >= 2.1.0
   • Python packages: pip install -r requirements.txt

Dataset Preparation

Unzip and move dataset into ROOT/dataset or ROOT/data

Directory structure of dataset

    ├── data                
    │   ├── misc              
    │   ├── MSIs
    |   ├── Videos
    |   dc.tif
    |   PaviaU.mat          
    ├── dataset
    │   ├── bunny         
    │   ├── shapeNet

Run and test

• Inpainting: ./Inpainting.sh
• Denoising: ./Denoising.sh
• Upsampling ./Upsampling.sh

Acknowledgement

This implementation is based on / inspired by:

• reproducible-tensor-completion-state-of-the-art
• M $^2$ DMT
• HLRTF
• Continuous-Tensor-Toolbox
• DeepTensor
• SAPCU
• NeuralTPS
• NeuralPoints
• Grad-PU