GRASS: Generative Recursive Autoencoders for Shape Structures
GRASS: Generative Recursive Autoencoders for Shape Structures
复制标题
GRASS：形状结构的生成递归自动编码器
DOI：
10.1145/3072959.3073637
复制
发表时间：
2017-07-01
影响因子：
6.2
通讯作者：
Guibas, Leonidas
中科院分区：
文献类型：
--
作者：
Li, Jun;Xu, Kai;Guibas, Leonidas
我们介绍了一种新的神经网络体系结构，用于编码和合成3D形状，特别是它们的结构。我们的主要见解是，3D形状的有效特征是它们对部件的分层组织，这反映了基本的形状内关系，如邻接和对称。我们开发了一个基于递归神经网络(RvNN)的自动编码器，将平面的、未标记的、任意的零件布局映射到紧凑的代码。该代码有效地捕获了人造3D对象的层次结构，尽管是固定维度的，但结构复杂程度各不相同：相关的解码器将代码映射回完整的层次结构。使用对抗性设置进一步调谐所学习的双向映射以产生合理结构的生成模型，从中可以采样新的结构。最后，我们的结构合成框架通过第二个经过训练的模块来扩展，该模块根据全局和局部结构环境生成细粒度的零件几何图形，从而产生完整的3D形状生成管道。我们证明，在没有监督的情况下，我们的网络遵循感知分组原则学习有意义的结构层次，产生能够实现形状分类和部分匹配等应用的紧凑代码，并支持在拓扑和几何上有显著变化的形状合成和内插。
We introduce a novel neural network architecture for encoding and synthesis of 3D shapes, particularly their structures. Our key insight is that 3D shapes are effectively characterized by their hierarchical organization of parts, which reflects fundamental intra-shape relationships such as adjacency and symmetry. We develop a recursive neural net (RvNN) based autoencoder to map a flat, unlabeled, arbitrary part layout to a compact code. The code effectively captures hierarchical structures of man-made 3D objects of varying structural complexities despite being fixed-dimensional: an associated decoder maps a code back to a full hierarchy. The learned bidirectional mapping is further tuned using an adversarial setup to yield a generative model of plausible structures, from which novel structures can be sampled. Finally, our structure synthesis framework is augmented by a second trained module that produces fine-grained part geometry, conditioned on global and local structural context, leading to a full generative pipeline for 3D shapes. We demonstrate that without supervision, our network learns meaningful structural hierarchies adhering to perceptual grouping principles, produces compact codes which enable applications such as shape classification and partial matching, and supports shape synthesis and interpolation with significant variations in topology and geometry.