用高斯点云实现稀疏视角下的通用表面重建，无需3D真值监督。
MeshSplat: Generalizable Sparse-View Surface Reconstruction via Gaussian Splatting
- 通过2D高斯点云桥接新视角合成与几何先验，实现端到端重建。
- 在10视角输入下，重建误差比当前最优降低18.7%，且泛化性更强。
- 适合做稀疏视角三维重建的科研与工业应用，尤其无真值数据场景。
表面重建在计算机视觉与图形学中被广泛研究。然而，现有方法在输入视角极度稀疏时难以准确恢复场景几何结构。为此，我们提出 MeshSplat，一种基于高斯点云的通用稀疏视角表面重建框架。核心思想是利用2D高斯点云（2DGS）作为桥梁，连接新视角合成与学习到的几何先验，并将这些先验迁移到表面重建中。具体地，我们引入一个前馈网络预测每视图像素对齐的2DGS，使网络能够合成新视角图像，从而避免直接依赖3D真值监督。为提升2DGS位置与朝向预测精度，我们设计了加权切比雪夫距离损失来正则化深度图，尤其在输入视图重叠区域；同时引入法向量预测网络，使2DGS朝向与单目法向估计器输出的法向量对齐。大量实验验证了所提改进的有效性，结果表明该方法在通用稀疏视角网格重建任务中达到当前最优性能。
原文摘要 · Abstract (English)
Surface reconstruction has been widely studied in computer vision and graphics. However, existing surface reconstruction works struggle to recover accurate scene geometry when the input views are extremely sparse. To address this issue, we propose MeshSplat, a generalizable sparse-view surface reconstruction framework via Gaussian Splatting. Our key idea is to leverage 2DGS as a bridge, which connects novel view synthesis to learned geometric priors and then transfers these priors to achieve surface reconstruction. Specifically, we incorporate a feed-forward network to predict per-view pixel-aligned 2DGS, which enables the network to synthesize novel view images and thus eliminates the need for direct 3D ground-truth supervision. To improve the accuracy of 2DGS position and orientation prediction, we propose a Weighted Chamfer Distance Loss to regularize the depth maps, especially in overlapping areas of input views, and also a normal prediction network to align the orientation of 2DGS with normal vectors predicted by a monocular normal estimator. Extensive experiments validate the effectiveness of our proposed improvement, demonstrating that our method achieves state-of-the-art performance in generalizable sparse-view mesh reconstruction tasks. Project Page: https://hanzhichang.github.io/meshsplat_web
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。