Lingo improves sparse view x ray ct images with faster training
LINGO: Latent Initialization and Gradient Optimization for Sparse-view X-ray Novel View Synthesis and CT Reconstruction with 3D Gaussian Splatting
Computer Vision and Pattern RecognitionGraphics
Summary
Taking X-ray images from only a few angles makes it hard to get clear pictures of inside objects or bodies because the images can be noisy and unclear. The authors introduced LINGO, a new method that smartly starts with better guesses about the shape and uses special adjustments during training to make clearer 3D pictures faster. Their method works by reducing background noise and improving details in areas that are usually hard to see. Testing shows LINGO creates sharper images more quickly than older approaches when using limited X-ray views.
What this means in practice
- •For medical imaging teams: Enhance CT reconstructions from few X-ray views to reduce scan times and radiation exposure while improving image quality.
- •For industrial nondestructive testing teams: Generate clearer 3D images of objects using fewer X-ray angles to speed inspections and identify flaws more reliably.
Authors
Lifeng Xing, Dequan Jin, Kunpeng Bu, Peigeng He, Shihui Ying
Abstract
In novel view synthesis and Computed Tomography (CT) reconstruction with sparse-view X-ray imaging, insufficient angular coverage leads to structural ambiguity and accumulated noise. Integrating 3D Gaussian Splatting (3DGS) with X-ray absorption physics can achieve promising results, but it suffers from noisy initialization, positional insensitivity, and weak gradients in low-density regions. In this paper, we propose a unified Latent Initialization and Gradient Optimization (LINGO) framework to address these issues. LINGO combines latent mask-space initialization with dynamic gradient optimization to improve point cloud structural completeness while accelerating training. It constructs voxel-level 3D filters from X-ray masks to robustly suppress background noise and provide reliable geometric priors. By employing an adaptive voxel scaling strategy and dynamically scaling loss, LINGO can adjust spatial resolution and explicitly amplify gradients in low-density structures. To evaluate the quality of initialization, we introduce the Initialization Point Cloud Structural Deviation (IPSD) metric. Experiments on the X3D dataset indicate that for the novel view synthesis task, LINGO improves the Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM) by an average of 0.72 and 0.0039, respectively, over baselines under identical sparse-view settings, achieving comparable reconstruction quality within 5k steps to state-of-the-art models typically trained with 30k iterations. For the CT reconstruction task, LINGO also demonstrates consistent improvements, with average PSNR and SSIM gains of 0.36 and 0.0134. These results highlight LINGO's effectiveness in both accelerating training and enhancing reconstruction quality across different sparse-view imaging scenarios.