LATO.2：基于顶点和拓扑流的因式分解3D网格生成
LATO.2: Factorized 3D Mesh Generation with Vertex and Topology Flow
- Huazhong University of Science and Technology（华中科技大学）
- Meshy AI
- Technical University of Munich（慕尼黑技术大学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
研究针对现有拓扑感知网格生成方法的不足，提出LATO.2框架，将网格生成分解为顶点流和连通性流，由专用变分自编码器支持，具有逐部分生成和拓扑自适应编辑优势，实验证明其超越现有方法。
AI中文摘要：
最近，在精心设计的潜在表示上进行流匹配已成为拓扑感知网格生成的强大范例。然而，现有方法在联合潜在空间中联合对顶点和连通性进行建模，将连续的顶点几何与离散的组合结构纠缠在一起，这使流学习变得复杂，并表现为顶点漂移和表面破裂。我们提出了LATO.2，这是一个因式分解流匹配框架，它将网格生成分解为一个顶点流，然后是一个基于已实现顶点的连通性流，两个阶段都锚定在一个共享的粗体素支架上。专用变分自编码器支持这两个阶段，以亚体素精度恢复顶点，并将离散连通性嵌入连续潜在空间。我们展示了这种因式分解的两个独特优势：（i）逐部分生成，其中支架被分割，每个部分以全潜在容量合成，产生比整体潜在更高分辨率的网格；（ii）拓扑自适应编辑，其中操纵第一阶段顶点会诱导相应的连通性，而无需重新优化。实验表明，LATO.2在几何保真度和连通性质量方面超过了现有的拓扑感知网格生成器。
英文摘要：
Flow matching over carefully designed latent representations has recently emerged as a powerful paradigm for topology-aware mesh generation. Existing approaches, however, model vertices and connectivity jointly in a joint latent space, entangling continuous vertex geometry with discrete combinatorial structure; this complicates flow learning and manifests as drifting vertices and broken surfaces. We present LATO.2, a factorized flow matching framework that decomposes mesh generation into a vertex flow followed by a connectivity flow conditioned on the realized vertices, with both stages anchored to a shared coarse voxel scaffold. Dedicated VAEs underpin the two stages, recovering vertices at sub-voxel precision and embedding discrete connectivity into a continuous latent space. We demonstrate two advantages unique to this factorization: (i) part-wise generation, in which the scaffold is partitioned and each part synthesized at full latent capacity, yielding substantially higher-resolution meshes than a monolithic latent permits; and (ii) topology-adaptive editing, in which manipulating first-stage vertices induces the corresponding connectivity without re-optimization. Experiments show that LATO.2 surpasses state-of-the-art topology-aware mesh generators in geometric fidelity and connectivity quality.
