TR-GS: High-Fidelity Sparse-View CT Volumetric Rendering via t-Distribution Gaussian Splatting and Ray-Confidence Modeling
2026-08-17 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors developed a new method called TR-GS to improve 3D images from CT scans taken with fewer X-ray views, which helps reduce radiation exposure. Their approach uses a special type of mathematical shape (Student's t-distribution) instead of the usual ones to better handle missing or unclear data. They also add a system to judge how reliable each X-ray path is and use that to clean up noise while keeping important details. Tests on both made-up and real data showed their method works better or similarly to existing ones. This can help doctors see clearer 3D images for diagnosis and treatment planning.
Sparse-view CT3D Gaussian SplattingStudent's t-distributionVolumetric renderingRay-confidence modelWavelet regularizationRadiation exposureStructural artifactsMedical visualizationExtended Reality (XR)
Authors
Zedong Xiao, Yiren Wang, Zhou Liu, Xiaolin Liu, Zhangji Lu
Abstract
High-fidelity 3D medical visualization supports applications such as clinical assessment and surgical planning. Sparse-view computed tomography (CT) can reduce projection requirements and associated radiation exposure, but limited observations may introduce structural artifacts and reconstruction uncertainty. Although 3D Gaussian Splatting (3DGS) provides an efficient explicit representation for volumetric rendering, existing CT methods based on standard Gaussian primitives may be sensitive to unreliable observations under sparse-view acquisition. We present TR-GS, a Gaussian-splatting framework for sparse view CT volumetric rendering. TR-GS replaces standard Gaussian primitives with projectable Student's t-distribution primitives and introduces a ray-confidence model that regulates their degrees of freedom according to local ray observability. Confidence-guided 3D wavelet regularization is further used to balance high-frequency detail preservation and noise suppression. This work is licensed under a Creative Commons Attribution 4.0 International License. Experiments on synthetic and real-world datasets show that TR-GS improves over representative baselines in most evaluated settings and remains competitive in the remaining cases. The resulting volumetric representations may support downstream medical multimedia applications, including XR-based visualization and interactive clinical rendering.