SPyRiT is a PyTorch-based deep image reconstruction package primarily designed for single-pixel imaging.
The spyrit package is available for Linux, MacOs and Windows. We recommend to use a virtual environment.
(user mode)
pip install spyrit
(developper mode)
git clone https://github.com/openspyrit/spyrit.git
cd spyrit
pip install -e .
On Windows you may need to install PyTorch first. It may also be necessary to run the following commands using administrator rights (e.g., starting your Python environment with administrator rights).
Adapt the two examples below to your configuration (see here for the latest instructions)
(CPU version using pip)
pip3 install torch torchvision torchaudio
(GPU version using conda)
conda install pytorch torchvision torchaudio pytorch-cuda=12.4 -c pytorch -c nvidiaThen, install SPyRiT using pip:
(user mode)
pip install spyrit
(developper mode)
git clone https://github.com/openspyrit/spyrit.git
cd spyrit
pip install -e .
To check the installation, run in your python terminal:
import spyrit
To start, check the documentation tutorials. These tutorials must be runned from tutorial folder (they load image samples from spyrit/images/):
cd spyrit/tutorial/
More advanced reconstruction examples can be found in spyrit-examples/tutorial. Run advanced tutorial in colab:
https://spyrit.readthedocs.io/
- Juan Abascal - Website
- Thomas Baudier
- Sebastien Crombez
- Nicolas Ducros - Website
- Antonio Tomas Lorente Mur - Website
- Romain Phan
- Fadoua Taia-Alaoui
When using SPyRiT in scientific publications, please cite the following paper:
- G. Beneti-Martin, L Mahieu-Williame, T Baudier, N Ducros, "OpenSpyrit: an Ecosystem for Reproducible Single-Pixel Hyperspectral Imaging," Optics Express, Vol. 31, No. 10, (2023). https://doi.org/10.1364/OE.483937.
When using SPyRiT specifically for the denoised completion network, please cite the following paper:
- A Lorente Mur, P Leclerc, F Peyrin, and N Ducros, "Single-pixel image reconstruction from experimental data using neural networks," Opt. Express 29, 17097-17110 (2021). https://doi.org/10.1364/OE.424228.
This project is licensed under the LGPL-3.0 license - see the LICENSE.md file for details
- Jin LI for his implementation of Convolutional Gated Recurrent Units for PyTorch
- Erik Lindernoren for his processing of the UCF-101 Dataset.