Meshes are important representations of physical 3D entities in the virtual world. Applications like rendering, simulations and 3D printing require meshes to be manifold so that they can interact with the world like the real objects they represent. Prior methods generate meshes with great geometric accuracy but poor manifoldness. In this work, we propose Neural Mesh Flow (NMF) to generate two-manifold meshes for genus-0 shapes. Specifically, NMF is a shape auto-encoder consisting of several Neural Ordinary Differential Equation (NODE)[1] blocks that learn accurate mesh geometry by progressively deforming a spherical mesh. Training NMF is simpler compared to state-of-the-art methods since it does not require any explicit mesh-based regularization. Our experiments demonstrate that NMF facilitates several applications such as single-view mesh reconstruction, global shape parameterization, texture mapping, shape deformation and correspondence. Importantly, we demonstrate that manifold meshes generated using NMF are better-suited for physically-based rendering and simulation. Code and data are released.
翻译:3D 实体在虚拟世界中的物理 3D 实体 重要 。 3D 实体在虚拟世界中的物理 3D 代表 。 诸如 模拟、 模拟 和 3D 打印 等应用要求 meshe 具有多重性, 以便他们能够像真实的物体一样与世界互动 。 先前的方法生成的 meshe 具有很高的几何精确的几何 。 先前的方法生成的 meshe 精度却不甚高 。 在这项工作中, 我们提议神经气象流( NMF ) 来生成 genus- 0 形状的双维的模模项。 具体地说, NMF 是一个形状的自动编码, 由多个神经普通差异化( NODE) [1] 组成, 通过逐渐变形来学习精确的网格几块 。 培训NMF 要比最先进的方法更简单, 因为不需要任何明确的网格规范 。 我们的实验表明 NMF 能够促进多种应用, 如单视网目图的重建、 全球 参数参数、 绘图、 绘图、 、 形状绘图、 变形和对等和对等 。 我们证明 数据 。 。 。 我们证明使用 NMF 的多元和 。