The Flow has not summarised this story yet — read it at Hugging Face Trending Papers.
SparseTokenGS is a Gaussian-based, feed-forward method for sparse-view cone-beam CT reconstruction that uses a reusable token pool to maintain scan-wide context during decoding. Anchor Token Fusion and Point-Token Attention query this shared pool at anchors and query coordinates, preserving local Gaussian priors while enabling reconstruction in a single forward pass without per-scan optimization. Experiments on chest, dental, volcanic rock, and chicken datasets show that SparseTokenGS achieves superior reconstruction quality, achieving 29.51 dB PSNR and 0.842 SSIM from only six dental projections and maintaining separable tooth-and-pulp structures for root‑canal localization.
Generating a 3D dental volume from a single panoramic radiograph (PXR) could provide a low-radiation alternative to Cone-Beam Computed Tomography (CBCT), but the problem is highly underdetermined: panoramic acquisition integrates 3D attenuation along curved X-ray paths into a 2D image, leaving depth-resolved anatomy unobserved. Existing implicit and generative approaches often produce oversmoothed geometry or anatomically inconsistent hallucinations, lacking geometry-driven supervision and relying on smooth representations unable to precisely localize sharp anatomical boundaries.
The paper presents a three‑stage framework to extend the field of view in dental cone‑beam CT scans that are limited by truncated detector configurations. First, an implicit neural representation estimates missing projection data; second, an iterative reconstruction generates a secondary volumetric image with better anatomical consistency; third, a fast diffusion model refines the image for enhanced quality. Experiments show reduced truncation artifacts and improved reconstruction of structures beyond the original field of view.
Sparse-view computed tomography is a severely ill-posed inverse problem, where recent 3D Gaussian Splatting methods offer an efficient explicit representation for tomographic reconstruction. However, we find that projection-domain optimization can be misleading in this setting: the rendered projections may continue to improve while the reconstructed volume deteriorates.
Dental cone-beam computed tomography (CBCT) systems often employ detector configurations that provide a truncated field of view (FOV) that only captures a small part of the patient's anatomy. In this...
arXiv:2609.37605v1 Announce Type: cross Abstract: Supervised deep learning has advanced sparse-view tomographic reconstruction. However, conventional models, which typically map filtered back-project...