DiffNR: Diffusion-Enhanced Neural Representation Optimization for Sparse-View 3D Tomographic Reconstruction

2026-04-23

AI Summary

This work addresses the severe artifacts commonly arising in neural representations for sparse-view CT reconstruction by proposing DiffNR, a novel framework that incorporates diffusion priors to enhance the optimization process. DiffNR periodically employs a single-step diffusion model, SliceFixer, to generate pseudo-reference volumetric data, providing three-dimensional-aware supervision for under-constrained regions. A repair-and-enhance strategy is introduced to balance reconstruction quality and computational efficiency by minimizing frequent calls to the diffusion model. By integrating neural representations—such as neural radiance fields and 3D Gaussians—with tailored conditional layers and dataset-specific fine-tuning, DiffNR achieves an average PSNR improvement of 3.99 dB across multiple datasets, significantly enhancing reconstruction fidelity while demonstrating strong cross-domain generalization.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Search and Optimization: Non-convex Optimization
• Machine Learning: Representation Learning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Neural representations (NRs), such as neural fields and 3D Gaussians, effectively model volumetric data in computed tomography (CT) but suffer from severe artifacts under sparse-view settings. To address this, we propose DiffNR, a novel framework that enhances NR optimization with diffusion priors. At its core is SliceFixer, a single-step diffusion model designed to correct artifacts in degraded slices. We integrate specialized conditioning layers into the network and develop tailored data curation strategies to support model finetuning. During reconstruction, SliceFixer periodically generates pseudo-reference volumes, providing auxiliary 3D perceptual supervision to fix underconstrained regions. Compared to prior methods that embed CT solvers into time-consuming iterative denoising, our repair-and-augment strategy avoids frequent diffusion model queries, leading to better runtime performance. Extensive experiments show that DiffNR improves PSNR by 3.99 dB on average, generalizes well across domains, and maintains efficient optimization.

Problem

• sparse-view
• 3D tomographic reconstruction
• neural representations
• artifacts
• computed tomography

Innovation

• Diffusion prior
• Neural representation
• Sparse-view CT
• SliceFixer
• 3D tomographic reconstruction