Liang Jiye, Du Zijin, Liang Jianqing, Yao Kaixuan, Cao Feilong
IEEE Trans Pattern Anal Mach Intell. 2023 Dec;45(12):14975-14989. doi: 10.1109/TPAMI.2023.3298711. Epub 2023 Nov 3.
Graph convolutional neural networks can effectively process geometric data and thus have been successfully used in point cloud data representation. However, existing graph-based methods usually adopt the K-nearest neighbor (KNN) algorithm to construct graphs, which may not be optimal for point cloud analysis tasks, owning to the solution of KNN is independent of network training. In this paper, we propose a novel graph structure learning convolutional neural network (GSLCN) for multiple point cloud analysis tasks. The fundamental concept is to propose a general graph structure learning architecture (GSL) that builds long-range and short-range dependency graphs. To learn optimal graphs that best serve to extract local features and investigate global contextual information, respectively, we integrated the GSL with the designed graph convolution operator under a unified framework. Furthermore, we design the graph structure losses with some prior knowledge to guide graph learning during network training. The main benefit is that given labels and prior knowledge are taken into account in GSLCN, providing useful supervised information to build graphs and thus facilitating the graph convolution operation for the point cloud. Experimental results on challenging benchmarks demonstrate that the proposed framework achieves excellent performance for point cloud classification, part segmentation, and semantic segmentation.
图卷积神经网络能够有效地处理几何数据，因此已成功应用于点云数据表示。然而，现有的基于图的方法通常采用K近邻（KNN）算法来构建图，由于KNN的解决方案与网络训练无关，这对于点云分析任务可能不是最优的。在本文中，我们提出了一种用于多点云分析任务的新型图结构学习卷积神经网络（GSLCN）。其基本概念是提出一种通用的图结构学习架构（GSL），用于构建长程和短程依赖图。为了分别学习最适合提取局部特征和研究全局上下文信息的最优图，我们在统一框架下将GSL与设计的图卷积算子相结合。此外，我们利用一些先验知识设计图结构损失，以在网络训练期间指导图学习。主要优点是GSLCN考虑了给定的标签和先验知识，为构建图提供了有用的监督信息，从而便于对点云进行图卷积操作。在具有挑战性的基准上的实验结果表明，所提出的框架在点云分类、部分分割和语义分割方面取得了优异的性能。
