Convolution on Rotation-Invariant and Multi-Scale Feature Graph for 3D Point Set Segmentation
Convolution on Rotation-Invariant and Multi-Scale Feature Graph for 3D Point Set Segmentation
复制标题
DOI：
10.1109/access.2020.3012613
复制
发表时间：
2020-07
期刊：
影响因子：
3.9
通讯作者：
T. Furuya;X. Hang;Ryutarou Ohbuchi;Jinliang Yao
中科院分区：
文献类型：
--
作者：
T. Furuya;X. Hang;Ryutarou Ohbuchi;Jinliang Yao
三维物体的抗旋转不变性是三维形状分析的基本性质之一。最近提出的算法通过使用固有的旋转不变的3D形状特征，即3D点之间的距离和角度作为深度神经网络(DNN)的输入，实现了旋转不变的3D点集分析。DNN捕捉几何特征之间的空间层次和上下文，以产生准确的分析结果。在本文中，我们进一步深入研究了基于DNN的旋转不变和高精度的3D点集分析方法。其中，3D点集分割是3D点集分析中最具挑战性的任务之一。我们提出了一种新的用于3D点集分割的DNN，称为旋转不变多尺度特征图卷积神经网络。我们的RMGnet比以前的方法更灵活，因为它接受任何手工制作的具有旋转不变性的3D形状特征作为输入。此外，为了准确地分割由不同尺寸的零件组成的3D点集，我们将手工特征提取的尺度随机化，并使用DNN对特征进行多分辨率分析。实验结果表明，该算法具有较高的分割精度和旋转不变性。
Invariance against rotation of 3D objects is one of the essential properties for 3D shape analysis. Recently proposed algorithms have achieved rotationally invariant 3D point set analysis by using inherently rotation-invariant 3D shape features, i.e., distances and angles among 3D points, as input to Deep Neural Networks (DNNs). The DNNs capture spatial hierarchy and context among the geometric features to produce accurate analytical results. In this article, we delve further into the DNN-based approach to rotation-invariant and highly accurate 3D point set analysis. In particular, we focus our attention on segmentation of 3D point sets, which is one of the most challenging among 3D point set analysis tasks. We propose a novel DNN for 3D point set segmentation called Rotation-invariant and Multi-scale feature Graph convolutional neural network, or RMGnet. Our RMGnet is more flexible than the previous methods as it accepts as input any handcrafted 3D shape features having rotation invariance. In addition, to accurately segment 3D point sets composed of parts having various sizes, we randomize scales at which handcrafted features are extracted and perform multi-resolution analysis of the features by using the DNN. Experimental evaluation demonstrates high segmentation accuracy as well as rotation invariance of the proposed RMGnet.