Semantic Point Cloud Segmentation Using Fast Deep Neural Network and DCRF
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...
Saved in:
Published in | Sensors (Basel, Switzerland) Vol. 21; no. 8; p. 2731 |
---|---|
Main Authors | , , , , , |
Format | Journal Article |
Language | English |
Published |
Switzerland
MDPI AG
13.04.2021
MDPI |
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | 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. |
---|---|
Bibliography: | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 This manuscript is extension version fo the conference paper: Yunbo Rao, Menghan Zhang, Zhanglin Cheng, Junmin Xue, Jiansu Pu, and Zairong Wang, Fast 3D Point Cloud Segmentation Using Deep Neural Network. In Proceeding of the IEEE International Conference on Internet of Things and Intelligent Applications (ITIA2020), Zhenjiang, China, 27–29 November 2020. |
ISSN: | 1424-8220 1424-8220 |
DOI: | 10.3390/s21082731 |