AirSplat：基于鲁棒前馈3D高斯散射的对齐与评分
AirSplat: Alignment and Rating for Robust Feed-Forward 3D Gaussian Splatting
AI总结：
本文提出AirSplat框架，通过自一致性姿态对齐和基于评分的透明度匹配技术，提升无姿态视角合成的重建质量。
摘要：
尽管3D视觉基础模型在视觉几何估计中表现出色，但其直接应用于通用视角合成仍具挑战性。本文提出AirSplat，一种新的训练框架，有效将3DVFMs的鲁棒几何先验转化为高保真无姿态视角合成。我们的方法引入了两个关键技术贡献：(1) 自一致性姿态对齐（SCPA），一种训练时反馈循环，确保像素对齐监督以解决姿态-几何不一致；(2) 基于评分的透明度匹配（ROM），利用稀疏视角视角合成教师模型的局部3D几何一致性知识来过滤劣化的基元。在大规模基准测试中，我们的方法在重建质量上显著优于最先进的无姿态视角合成方法。AirSplat展示了将3DVFMs适应于同时进行视觉几何估计和高质量视角合成的潜力。
英文摘要：
While 3D Vision Foundation Models (3DVFMs) have demonstrated remarkable zero-shot capabilities in visual geometry estimation, their direct application to generalizable novel view synthesis (NVS) remains challenging. In this paper, we propose AirSplat, a novel training framework that effectively adapts the robust geometric priors of 3DVFMs into high-fidelity, pose-free NVS. Our approach introduces two key technical contributions: (1) Self-Consistent Pose Alignment (SCPA), a training-time feedback loop that ensures pixel-aligned supervision to resolve pose-geometry discrepancy; and (2) Rating-based Opacity Matching (ROM), which leverages the local 3D geometry consistency knowledge from a sparse-view NVS teacher model to filter out degraded primitives. Experimental results on large-scale benchmarks demonstrate that our method significantly outperforms state-of-the-art pose-free NVS approaches in reconstruction quality. Our AirSplat highlights the potential of adapting 3DVFMs to enable simultaneous visual geometry estimation and high-quality view synthesis.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
