用单目图像重建高质量3D网格，结合高斯点与神经SDF提升精度。
MonoGSDF: Exploring Monocular Geometric Cues for Gaussian Splatting-Guided Implicit Surface Reconstruction
- 用神经SDF引导高斯点分布，实现更精确的表面建模
- 无需内存消耗大的Marching Cubes，推理效率更高
- 适用于任意尺度场景，适合真实世界单目重建任务
从单目图像准确重建三维网格仍是三维视觉中的关键挑战。尽管当前最先进的3D高斯溅射（3DGS）方法在基于光栅化渲染下能合成逼真的新视角图像，但其依赖稀疏显式原语严重限制了恢复封闭且拓扑一致三维表面的能力。我们提出MonoGSDF，一种将基于高斯的原语与神经符号距离场（SDF）相结合的新方法，以实现高质量重建。训练时，SDF指导高斯点的空间分布；推理时，高斯点作为先验重建表面，无需内存密集的Marching Cubes。为应对任意尺度场景，我们提出一种缩放策略以增强泛化能力。多分辨率训练方案进一步优化细节，来自现成估计器的单目几何线索也提升了重建质量。在真实数据集上的实验表明，MonoGSDF优于现有方法，同时保持高效。
原文摘要 · Abstract (English)
Accurate meshing from monocular images remains a key challenge in 3D vision. While state-of-the-art 3D Gaussian Splatting (3DGS) methods excel at synthesizing photorealistic novel views through rasterization-based rendering, their reliance on sparse, explicit primitives severely limits their ability to recover watertight and topologically consistent 3D surfaces.We introduce MonoGSDF, a novel method that couples Gaussian-based primitives with a neural Signed Distance Field (SDF) for high-quality reconstruction. During training, the SDF guides Gaussians' spatial distribution, while at inference, Gaussians serve as priors to reconstruct surfaces, eliminating the need for memory-intensive Marching Cubes. To handle arbitrary-scale scenes, we propose a scaling strategy for robust generalization. A multi-resolution training scheme further refines details and monocular geometric cues from off-the-shelf estimators enhance reconstruction quality. Experiments on real-world datasets show MonoGSDF outperforms prior methods while maintaining efficiency.
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。