TopoMesh: 通过拓扑统一实现高保真的网格自编码
TopoMesh: High-Fidelity Mesh Autoencoding via Topological Unification
AI总结：
TopoMesh通过拓扑统一框架解决网格生成中的拓扑不匹配问题，利用稀疏体素VAE实现高保真重建，提升几何细节和锐利特征的保留能力。
摘要：
高保真3D生成主流依赖于VAE-扩散管道，其中VAE的重建能力设定了生成质量的上限。现有VAE的核心挑战是真实网格与网络预测之间的表示不匹配：真实网格具有任意变化的拓扑，而VAE通常预测固定结构的隐式场（例如正则网格上的SDF）。这种固有不匹配阻止了显式网格级对应关系的建立，迫使现有工作依赖间接监督信号如SDF或渲染损失。因此，细几何细节，特别是锐利特征，在重建中表现不佳。为此，我们引入TopoMesh，一种基于稀疏体素的VAE，通过共享的双Marching Cubes（DMC）拓扑框架统一真实和预测的网格。具体而言，我们通过一种保留锐利边的算法将任意输入网格转换为DMC兼容的表示，使用L∞距离度量。我们的解码器输出相同DMC格式的网格，确保预测和目标网格具有相同的拓扑结构。这建立了顶点和面级别的显式对应关系，允许我们推导出显式的网格级监督信号，用于拓扑、顶点位置和面方向，具有清晰的梯度。我们的稀疏VAE架构利用这一统一框架，并通过教师强制和逐步分辨率训练进行训练，以实现稳定和高效的收敛。广泛的实验表明，TopoMesh在重建保真度上显著优于现有VAE，实现了对锐利特征和几何细节的优越保留。
英文摘要：
The dominant paradigm for high-fidelity 3D generation relies on a VAE-Diffusion pipeline, where the VAE's reconstruction capability sets a firm upper bound on generation quality. A fundamental challenge limiting existing VAEs is the representation mismatch between ground-truth meshes and network predictions: GT meshes have arbitrary, variable topology, while VAEs typically predict fixed-structure implicit fields (\eg, SDF on regular grids). This inherent misalignment prevents establishing explicit mesh-level correspondences, forcing prior work to rely on indirect supervision signals such as SDF or rendering losses. Consequently, fine geometric details, particularly sharp features, are poorly preserved during reconstruction. To address this, we introduce TopoMesh, a sparse voxel-based VAE that unifies both GT and predicted meshes under a shared Dual Marching Cubes (DMC) topological framework. Specifically, we convert arbitrary input meshes into DMC-compliant representations via a remeshing algorithm that preserves sharp edges using an L$\infty$ distance metric. Our decoder outputs meshes in the same DMC format, ensuring that both predicted and target meshes share identical topological structures. This establishes explicit correspondences at the vertex and face level, allowing us to derive explicit mesh-level supervision signals for topology, vertex positions, and face orientations with clear gradients. Our sparse VAE architecture employs this unified framework and is trained with Teacher Forcing and progressive resolution training for stable and efficient convergence. Extensive experiments demonstrate that TopoMesh significantly outperforms existing VAEs in reconstruction fidelity, achieving superior preservation of sharp features and geometric details.
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。
