2021-08-03

TopNet: Structural Point Cloud Decoder@TOC

TopNet: Structural Point Cloud Decoder 文章解读

TopNet: Structural Point Cloud Decoder

Abstract

3D point cloud generation is of great use for 3D scene modeling and understanding. Real-world 3D object point clouds can be properly described by a collection of lowlevel and high-level structures such as surfaces, geometric primitives, semantic parts, etc. In fact, there exist many different representations of a 3D object point cloud as a set of point groups. Existing frameworks for point cloud generation either do not consider structure in their proposed solutions, or assume and enforce a specific structure/topology, e.g. a collection of manifolds or surfaces, for the generated point cloud of a 3D object. In this work, we propose a novel decoder that generates a structured point cloud without assuming any specific structure or topology on the underlying point set. Our decoder is softly constrained to generate a point cloud following a hierarchical rooted tree structure. We show that given enough capacity and allowing for redundancies, the proposed decoder is very flexible and able to learn any arbitrary grouping of points including any topology on the point set. We evaluate our decoder on the task of point cloud generation for 3D point cloud shape completion. Combined with encoders from existing frameworks, we show that our proposed decoder significantly outperforms state-of-the-art 3D point cloud completion methods on the Shapenet dataset.

摘要

三维点云的生成对于三维场景的建模和理解非常有用。真实世界中的三维对象点云可以由一组低层和高层结构（如曲面、几何图元、语义部分等）正确描述。事实上，三维对象点云作为一组点组存在许多不同的表示形式。现有的点云生成框架要么不考虑它们提出的解决方案中的结构，要么假设和执行特定的结构/拓扑，例如，用于生成3D对象的点云的歧管或表面的集合。在这项工作中，我们提出了一种新的解码器，它可以生成结构化的点云，而不需要在基础点集上假设任何特定的结构或拓扑。我们的解码器是软约束生成一个点云以下的分层根树结构。我们证明，给定足够的容量并允许冗余，所提出的解码器非常灵活，能够学习任意分组的点，包括点集上的任何拓扑。我们评估了解码器的点云生成任务，以完成3D点云形状。结合现有框架中的编码器，我们证明了我们提出的解码器在Shapenet数据集上的性能明显优于最先进的3D点云完成方法。