SPyRiT 3.0: an open source package for single-pixel imaging based on deep learning

Single-pixel imaging is able to acquire an image from a few point measurements thanks to dedicated reconstruction algorithms. In recent years, reconstruction approaches based on deep learning have outperformed most alternatives. However, computational experiments and data-driven methods have become difficult, if not impossible, to reproduce. The development of tools enabling reproducibility and benchmarking is therefore now essential. This paper describes SPyRiT, an open-source PyTorch-based toolbox capable of handling various simulation configurations and reconstruction methods based on deep learning. In particular, SPyRiT implements both existing and new supervised and plug-and-play methods, including post-processing and iterative strategies. Our reconstruction results demonstrate that supervised methods trained on simulated data can be successfully applied to experimental data when the signal-to-noise ratio of the measurements is higher or equal to that of the training phase. On the other hand, the hyperparameter of the plug-and-play methods can be tuned to manage lower signal-to-noise ratios. Among the supervised methods, DC-Net is found to be robust to deviations in the noise level, achieving similar results to plug-and-play methods without hyperparameter selection, while offering low memory requirements and short reconstruction times. The modularity of SPyRiT enables the rigorous benchmarking of reconstructions based on deep learning in single-pixel imaging, as well as in related fields such as ghost imaging. Thanks to its modularity and versatility, SPyRiT is suitable for further studies beyond this work and could also benefit other modalities in the field of computational optics.