Neural Mesh Flow: 3D Manifold Mesh Generationvia Diffeomorphic Flows
Neural Mesh Flow: 3D Manifold Mesh Generationvia Diffeomorphic Flows
复制标题
DOI：
--
复制
发表时间：
2020-07
期刊：
影响因子：
--
通讯作者：
Kunal Gupta;Manmohan Chandraker
中科院分区：
文献类型：
--
作者：
Kunal Gupta;Manmohan Chandraker
网格是虚拟世界中三维实体的重要表示形式。像渲染、模拟和3D打印这样的应用需要网格是多种多样的，这样它们就可以像它们所代表的真实物体一样与世界互动。现有方法生成的网格几何精度高，但流形差。在这项工作中，我们提出了神经网格流（NMF）来生成属0形状的双流形网格。具体来说，NMF是一个形状自编码器，由几个神经常微分方程（NODE）[1]块组成，通过逐步变形球形网格来学习精确的网格几何形状。与最先进的方法相比，训练NMF更简单，因为它不需要任何显式的基于网格的正则化。实验表明，NMF有助于单视图网格重建、全局形状参数化、纹理映射、形状变形和对应等应用。重要的是，我们证明了使用NMF生成的流形网格更适合基于物理的渲染和模拟。代码和数据将会发布。
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 will be released.