TG-Field: Geometry-Aware Radiative Gaussian Fields for Tomographic Reconstruction

2026-02-12

AI Summary

This work addresses the challenge of severe artifacts in computed tomography (CT) reconstruction under highly sparse angular sampling and dynamic motion. To this end, we propose a geometry-aware Gaussian deformation reconstruction framework that integrates multi-resolution hash encoding, time-conditioned Gaussian representations, and a spatiotemporal attention mechanism. The method further incorporates geometric prior regularization and a respiratory motion flow network to effectively model both static anatomical structures and dynamic deformations. Extensive experiments on synthetic and real-world datasets demonstrate that our approach significantly outperforms existing methods, achieving high-fidelity and temporally consistent CT reconstructions even under ultra-sparse viewing conditions.

Technology Category

• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Computer Vision: 3D Computer Vision
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

3D Gaussian Splatting (3DGS) has revolutionized 3D scene representation with superior efficiency and quality. While recent adaptations for computed tomography (CT) show promise, they struggle with severe artifacts under highly sparse-view projections and dynamic motions. To address these challenges, we propose Tomographic Geometry Field (TG-Field), a geometry-aware Gaussian deformation framework tailored for both static and dynamic CT reconstruction. A multi-resolution hash encoder is employed to capture local spatial priors, regularizing primitive parameters under ultra-sparse settings. We further extend the framework to dynamic reconstruction by introducing time-conditioned representations and a spatiotemporal attention block to adaptively aggregate features, thereby resolving spatiotemporal ambiguities and enforcing temporal coherence. In addition, a motion-flow network models fine-grained respiratory motion to track local anatomical deformations. Extensive experiments on synthetic and real-world datasets demonstrate that TG-Field consistently outperforms existing methods, achieving state-of-the-art reconstruction accuracy under highly sparse-view conditions.

Problem

• sparse-view CT reconstruction
• dynamic motion artifacts
• 3D Gaussian Splatting
• tomographic reconstruction
• temporal coherence

Innovation

• 3D Gaussian Splatting
• tomographic reconstruction
• sparse-view CT
• spatiotemporal attention
• motion-flow network