This project is for fast-ish reconstruction using the concept of Gaussian process tomography (see Svensson, J., Li, Dong, eta al, and Wang, T., et al) while applying nonnegativity constraints. The motivation for this is the benefit that nonnegative constraints provide in the context of reconstruction from generalized detector readout. Please see arXiv:1912.01058 for a preprint article explaining the motivating use case.
- 2-D: launch in Google Colab
- 3-D: launch in Google Colab
- Coded: launch in Google Colab
- Hybrid Coded and Pixelated: launch in Google Colab