重新审视稀疏视角重建下基于泼溅的计算机断层成像中的位姿敏感性
Revisiting Pose Sensitivity in Splat-based Computed Tomography under Sparse-view Reconstruction
把 3D Gaussian splatting 的可微投影算子对采集几何位姿的敏感性讲清楚，并给出在重建中联合优化位姿的稳定梯度框架；做稀疏视角重建或位姿不精确场景（如 NVS、SLAM 式联合优化）的视觉研究者可直接借鉴其位姿扰动传播分析与轻量联合精修思路。
原文依据 we show that these artifacts primarily originate from pose inaccuracies in the acquisition geometry rather than from view sparsity itself
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论X 射线计算机断层成像（CT）通过让 X 射线穿透目标所获得的投影图像来重建物体的体表示。近期的基于泼溅的断层成像将体表示为 3D 高斯的连续分布，在锥束稀疏视角 CT 中同时展现出高重建质量与快速收敛。然而，当部署到视角数量有限且分布不均匀的真实 CT 系统中时，我们观察到明显的条纹与条带伪影，其严重程度远超传统重建方法。通过细致分析，我们表明这些伪影主要源于采集几何中的位姿不准确，而非视角稀疏性本身。我们重新审视了泼溅表述中的位姿敏感性，并推导出一个稳定的、基于梯度的框架，在重建过程中联合精修几何参数。我们的研究不仅刻画了位姿扰动如何通过可微投影算子传播，还揭示了基于泼溅的 CT 为何对几何失配尤为脆弱。所得表述保持轻量、易于集成到现有流程中，同时在真实稀疏视角条件下显著提升重建保真度。
英文原文摘要 1292 字符
X-ray computed tomography (CT) reconstructs volumetric representations of objects from projection images obtained by transmitting X-rays through a target. Recent splat-based tomography, which represents a volume as a continuous distribution of 3D Gaussians, has demonstrated both high reconstruction quality and fast convergence in cone-beam sparse-view CT. However, when deployed in real CT systems with limited and non-uniform view distributions, we observe distinctive streak and strip artifacts that are far more pronounced than in conventional reconstruction methods. Through detailed analysis, we show that these artifacts primarily originate from pose inaccuracies in the acquisition geometry rather than from view sparsity itself. We revisit pose sensitivity in the splatting formulation and derive a stable gradient-based framework that jointly refines geometric parameters during reconstruction. Our study not only identifies how pose perturbations propagate through the differentiable projection operator but also reveals why splat-based CT is particularly vulnerable to geometric misalignment. The resulting formulation remains lightweight and easily integrable into existing pipelines while substantially improving reconstruction fidelity under real-world sparse-view conditions.