LATO: 3D Mesh Flow Matching with Structured TOpology Preserving LAtents
📌 核心贡献
想象一下，不是直接在空气中凭空「画」一个3D模型，而是先有一个「骨架」或者说「橡皮泥团」，LATO就是发明了一种聪明的方法来描述这个橡皮泥团的「结构」，然后用新的捏合技巧（流匹配）去精准地塑造它，还能确保捏出来的所有连接和形状都非常符合原始骨架，不会出现奇形怪状的破损。
📖 摘要
In this paper, we introduce LATO, a novel topology-preserving latent representation that enables scalable, flow matching-based synthesis of explicit 3D meshes. LATO represents a mesh as a Vertex Displacement Field (VDF) anchored on surface, incorporating a sparse voxel Variational Autoencoder (VAE) to compress this explicit signal into a structured, topology-aware voxel latent. To decapsulate the mesh, the VAE decoder progressively subdivides and prunes latent voxels to instantiate precise vertex locations. In the end, a dedicated connection head queries the voxel latent to predict edge connectivity between vertex pairs directly, allowing mesh topology to be recovered without isosurface extraction or heuristic meshing. For generative modeling, LATO adopts a two-stage flow matching process, first synthesizing the structure voxels and subsequently refining the voxel-wise topology features. Compared to prior isosurface/triangle-based diffusion models and autoregressive generation approaches, LATO generates meshes with complex geometry, well-formed topology while being highly efficient in inference.
🔍 关键信息
| 字段 | 内容 |
|---|---|
| 发表 | 2026-03-06 |
| 机构 | LATO: 3D Mesh Flow Matching with Structured TOpology Preserving LAtents |
| 分类 | cs.CV |
| 链接 | arXiv |