In recent years, 3D point clouds (PCs) have gained significant attention due to their diverse applications across various fields such as computer vision (CV), condition monitoring, virtual reality, robotics, autonomous driving etc. Deep learning (DL) has proven effective in leveraging 3D PCs to address various challenges previously encountered in 2D vision. However, the application of deep neural networks (DNN) to process 3D PCs presents its own set of challenges. To address these challenges, numerous methods have been proposed. This paper provides an in-depth review of recent advancements in DL-based condition monitoring (CM) using 3D PCs, with a specific focus on defect shape classification and segmentation within industrial applications for operational and maintenance purposes. Recognizing the crucial role of these aspects in industrial maintenance, the paper provides insightful observations that offer perspectives on the strengths and limitations of the reviewed DL-based PC processing methods. This synthesis of knowledge aims to contribute to the understanding and enhancement of CM processes, particularly within the framework of remaining useful life (RUL), in industrial systems.

翻译：近年来，三维点云因其在计算机视觉、状态监测、虚拟现实、机器人技术、自动驾驶等多个领域的广泛应用而备受关注。深度学习已被证明能够有效利用三维点云解决先前在二维视觉中遇到的诸多挑战。然而，应用深度神经网络处理三维点云也带来了其自身的一系列问题。针对这些挑战，研究者提出了多种方法。本文深入综述了基于深度学习的三维点云状态监测领域的最新进展，重点关注工业应用中面向运维需求的缺陷形状分类与分割。基于对这些方面在工业维护中关键作用的认知，本文提供了富有洞察力的分析，揭示了所评述的基于深度学习的点云处理方法各自的优势与局限。这一知识综合旨在促进对状态监测过程的理解与改进，特别是在工业系统剩余使用寿命的框架下。