# Texture-preserving implicit neural representation for Cone beam CT truncated reconstruction

arXiv: https://arxiv.org/abs/2606.06039 · HTML: https://mlsanity.com/paper/2606.06039/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 24 of its 49 references that are in the map.
- First posted on arXiv 4 Jun 2026.

## Builds on (most cited first, 24 of 24)

- [3D Gaussian Splatting for Real-Time Radiance Field Rendering](https://mlsanity.com/paper/2308.04079/) (2023)
- [Instant Neural Graphics Primitives with a Multiresolution Hash Encoding](https://mlsanity.com/paper/2201.05989/) (2022)
- [Implicit Neural Representations with Periodic Activation Functions](https://mlsanity.com/paper/2006.09661/) (2020)
- [Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domains](https://mlsanity.com/paper/2006.10739/) (2020)
- [On the Spectral Bias of Neural Networks](https://mlsanity.com/paper/1806.08734/) (2018)
- [Beyond Periodicity: Towards a Unifying Framework for Activations in Coordinate-MLPs](https://mlsanity.com/paper/2111.15135/) (2021)
- [NAF: Neural Attenuation Fields for Sparse-View CBCT Reconstruction](https://mlsanity.com/paper/2209.14540/) (2022)
- [A Structured Dictionary Perspective on Implicit Neural Representations](https://mlsanity.com/paper/2112.01917/) (2021)
- [FINER: Flexible spectral-bias tuning in Implicit NEural Representation by Variable-periodic Activation Functions](https://mlsanity.com/paper/2312.02434/) (2023)
- [Structure-Aware Sparse-View X-ray 3D Reconstruction](https://mlsanity.com/paper/2311.10959/) (2023)
- [R2-Gaussian: Rectifying Radiative Gaussian Splatting for Tomographic Reconstruction](https://mlsanity.com/paper/2405.20693/) (2024)
- [MedNeRF: Medical Neural Radiance Fields for Reconstructing 3D-aware CT-Projections from a Single X-ray](https://mlsanity.com/paper/2202.01020/) (2022)
- [Neural Implicit k-Space for Binning-free Non-Cartesian Cardiac MR Imaging](https://mlsanity.com/paper/2212.08479/) (2022)
- [Disorder-invariant Implicit Neural Representation](https://mlsanity.com/paper/2304.00837/) (2023)
- [PIXEL: Physics-Informed Cell Representations for Fast and Accurate PDE Solvers](https://mlsanity.com/paper/2207.12800/) (2022)
- [PINs: Progressive Implicit Networks for Multi-Scale Neural Representations](https://mlsanity.com/paper/2202.04713/) (2022)
- [Polynomial Implicit Neural Representations For Large Diverse Datasets](https://mlsanity.com/paper/2303.11424/) (2023)
- [RHINO: Regularizing the Hash-based Implicit Neural Representation](https://mlsanity.com/paper/2309.12642/) (2023)
- [One Network to Solve All ROIs: Deep Learning CT for Any ROI using Differentiated Backprojection](https://mlsanity.com/paper/1810.00500/) (2018)
- [Field of View Extension in Computed Tomography Using Deep Learning Prior](https://mlsanity.com/paper/1911.01178/) (2019)
- [CT Field of View Extension Using Combined Channels Extension and Deep Learning Methods](https://mlsanity.com/paper/1908.09529/) (2019)
- [GR-Gaussian: Graph-Based Radiative Gaussian Splatting for Sparse-View CT Reconstruction](https://mlsanity.com/paper/2508.02408/) (2025)
- [Diffusion-based Generative Image Outpainting for Recovery of FOV-Truncated CT Images](https://mlsanity.com/paper/2406.04769/) (2024)
- [Towards the Spectral bias Alleviation by Normalizations in Coordinate Networks](https://mlsanity.com/paper/2407.17834/) (2024)

## Abstract (arXiv)

Cone-beam computed tomography (CBCT) frequently suffers from data truncation, which introduces severe artifacts and limits the effective field of view (FOV). Existing deep learning methods for truncated cone-beam computed tomography (CBCT) reconstruction suffer from serious limitations, including a strict reliance on supervised ground truth and a failure to account for continuous 3D spatial truncation variations. To address these challenges, we introduce a self-supervised 3D reconstruction framework based on neural scene representations. By directly mapping spatial coordinates to radiodensity under projection supervision, our approach inherently bypasses traditional filtering and backprojection operations, thereby fundamentally eliminating truncation-induced ring artifacts while enabling robust continuous 3D data extrapolation. However, coordinate networks are susceptible to an inherent spectral bias, which leads to a severe loss of clinically vital high-frequency textures. To resolve this bottleneck, we further incorporate a physics-based iterative refinement module into the neural scene representation architecture. Leveraging the artifact-free, extrapolated volume from the coordinate network as an optimal initialization, this module progressively re-extracts and injects high-frequency structural information from the original projections back into the volume. Extensive experiments on both simulated and real-world datasets demonstrate that our method successfully unifies the exceptional artifact suppression and extrapolation capabilities of neural networks with the high-fidelity detail preservation of iterative algorithms.