# GB-SVFBP: Gaussian-Based Shift-Variant FBP neural network

arXiv: https://arxiv.org/abs/2607.11584 · HTML: https://mlsanity.com/paper/2607.11584/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 4 of its 8 references that are in the map.
- First posted on arXiv 13 Jul 2026.

## Builds on (most cited first, 4 of 4)

- [PYRO-NN: Python Reconstruction Operators in Neural Networks](https://mlsanity.com/paper/1904.13342/) (2019)
- [Deep Learning Computed Tomography based on the Defrise and Clack Algorithm](https://mlsanity.com/paper/2403.00426/) (2024)
- [DRACO: Differentiable Reconstruction for Arbitrary CBCT Orbits](https://mlsanity.com/paper/2410.14900/) (2024)
- [Compressibility Analysis for the differentiable shift-variant Filtered Backprojection Model](https://mlsanity.com/paper/2501.11586/) (2025)

## Abstract (arXiv)

This paper proposes a Gaussian-Based Shift-Variant filtered backprojection (FBP) neural network, which is designed for the efficient reconstruction of non-circular trajectory cone beam computed tomography. The traditional differentiable shift-variant FBP model consists of a filtering component and a backprojection process. The filtering component includes operations such as weightings, differentiations, a 2D Radon transform, and a 2D backprojection. The proposed methods build on this framework by introducing a trainable 2D Gaussian model to represent the trajectory-related part in the filtering process, achieving a substantial reduction in the number of trainable parameters. Experimental results demonstrate that the proposed model reduces the parameter count by 99%, while only sacrificing a slight amount of reconstruction quality. Furthermore, the training time for each trajectory is reduced to one-fourth of the original, significantly accelerating convergence. These enhancements demonstrate a considerable augmentation in the model's practicality and effectiveness, making it a valuable asset for real-world applications.