School of Information and Software Engineering, University of Electronic Science and Technology of China, Chengdu 610054, China.
Yangtze Delta Region Institute (Huzhou), University of Electronic Science and Technology of China, Huzhou 313001, China.
Sensors (Basel). 2021 Apr 13;21(8):2731. doi: 10.3390/s21082731.
Accurate segmentation of entity categories is the critical step for 3D scene understanding. This paper presents a fast deep neural network model with Dense Conditional Random Field (DCRF) as a post-processing method, which can perform accurate semantic segmentation for 3D point cloud scene. On this basis, a compact but flexible framework is introduced for performing segmentation to the semantics of point clouds concurrently, contribute to more precise segmentation. Moreover, based on semantics labels, a novel DCRF model is elaborated to refine the result of segmentation. Besides, without any sacrifice to accuracy, we apply optimization to the original data of the point cloud, allowing the network to handle fewer data. In the experiment, our proposed method is conducted comprehensively through four evaluation indicators, proving the superiority of our method.
实体类别（Entity category）的精确分割（Segmentation）是三维场景理解（Scene understanding）的关键步骤。本文提出了一种快速的深度神经网络模型，该模型使用密集条件随机场（Dense Conditional Random Field，DCRF）作为后处理方法，可对三维点云场景进行精确的语义分割（Semantic segmentation）。在此基础上，引入了一个紧凑但灵活的框架（Framework），以并行执行点云的语义分割（Semantics of point clouds），有助于实现更精确的分割。此外，基于语义标签（Semantics labels），阐述了一种新颖的 DCRF 模型，以细化分割结果。此外，在不牺牲准确性的前提下，我们对点云的原始数据进行优化，使网络能够处理更少的数据。在实验中，我们通过四个评估指标对所提出的方法进行了全面的评估，证明了该方法的优越性。
