Invariance against rotations of 3D objects is an important property in analyzing 3D point set data. Conventional 3D point set DNNs having rotation invariance typically obtain accurate 3D shape features via supervised learning by using labeled 3D point sets as training samples. However, due to the rapid increase in 3D point set data and the high cost of labeling, a framework to learn rotation-invariant 3D shape features from numerous unlabeled 3D point sets is required. This paper proposes a novel self-supervised learning framework for acquiring accurate and rotation-invariant 3D point set features at object-level. Our proposed lightweight DNN architecture decomposes an input 3D point set into multiple global-scale regions, called tokens, that preserve the spatial layout of partial shapes composing the 3D object. We employ a self-attention mechanism to refine the tokens and aggregate them into an expressive rotation-invariant feature per 3D point set. Our DNN is effectively trained by using pseudo-labels generated by a self-distillation framework. To facilitate the learning of accurate features, we propose to combine multi-crop and cut-mix data augmentation techniques to diversify 3D point sets for training. Through a comprehensive evaluation, we empirically demonstrate that, (1) existing rotation-invariant DNN architectures designed for supervised learning do not necessarily learn accurate 3D shape features under a self-supervised learning scenario, and (2) our proposed algorithm learns rotation-invariant 3D point set features that are more accurate than those learned by existing algorithms. Code is available at https://github.com/takahikof/RIPT_SDMM

翻译：三维物体的旋转不变性是分析三维点集数据时的重要特性。具有旋转不变性的传统三维点集深度神经网络通常通过使用标注的三维点集作为训练样本，经由监督学习获得准确的三维形状特征。然而，由于三维点集数据的快速增长以及标注成本高昂，亟需一种能从大量未标注三维点集中学习旋转不变三维形状特征的框架。本文提出一种新颖的自监督学习框架，旨在目标层级上获取准确且旋转不变的三维点集特征。我们所提出的轻量级深度神经网络架构将输入的三维点集分解为多个全局尺度的区域（称为标记单元），这些标记单元保留了构成三维物体的局部形状的空间布局。我们利用自注意力机制来精炼这些标记单元，并将它们聚合成每个三维点集具有表现力的旋转不变特征。该深度神经网络通过自蒸馏框架生成的伪标签进行有效训练。为促进准确特征的学习，我们提出结合多裁剪与剪切混合数据增强技术，以增加训练用三维点集的多样性。通过全面评估，我们实证表明：(1) 为监督学习设计的现有旋转不变深度神经网络架构，在自监督学习场景下未必能学到准确的三维形状特征；(2) 我们提出的算法所学习的旋转不变三维点集特征比现有算法更为准确。代码见 https://github.com/takahikof/RIPT_SDMM