arXiv cs.AI· Linrui Dai, Jinxiu Liang, Chu Zhou, Mae Yamaguchi, Ryuya Takahashi, Imari Sato·· 14 小时前AI 评分29
基于姿态感知扩散先验的极稀疏视角CT重建：仅用2D投影训练
Extremely Sparse-View Computed Tomography from 2D Projections via Pose-Aware Diffusion Priors
AI 导读
研究提出从每训练样本仅一张2D投影学习姿态感知扩散先验，用于极稀疏视角CT重建，无需密集扫描的3D体数据。方法以测量视图保真、共享体重投影和投影物理一致性引导扩散采样，在1、4、8视角设置下于两个公开CT数据集取得最佳新视角合成分数，重建质量优于或媲美使用密集体监督的基线。
正文 · 原文
Abstract:Extremely sparse-view computed tomography (xSV-CT) is critical when scanning time, motion, or radiation tolerance limits each scan to only a few projections. In many such scenarios, the same acquisition limits also constrain dataset construction, making pre-reconstructed 3D volumes from dense scans unavailable. This exposes a gap in current sparse-view approaches: Optimization-based methods, which use the sparse measurements directly but lack data-driven priors to suppress artifacts, and learning-based methods, which use strong priors but usually require densely scanned training volumes acquired outside the xSV-CT regime. We study this train-sparse / infer-sparse setting by learning a pose-aware diffusion prior from one 2D projection per training sample, then guiding diffusion sampling with measured-view fidelity, shared-volume reprojection, and projection-physics consistency for tomographic reconstruction. On two public CT datasets across 1-, 4-, and 8-view settings, the proposed method achieves the best novel-view synthesis scores and best or competitive reconstruction quality compared with baselines that use dense-volume supervision.
| Comments: | NeurIPS 2026 poster |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.38319 [cs.CV] |
| (or arXiv:2610.38319v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2610.38319 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Linrui Dai [view email]
[v1]
Wed, 28 Oct 2026 06:08:39 UTC (31,238 KB)
来源：arXiv cs.AI · arxiv.org