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...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-04-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/21/8/2731 |
| _version_ | 1797537884763324416 |
|---|---|
| author | Yunbo Rao Menghan Zhang Zhanglin Cheng Junmin Xue Jiansu Pu Zairong Wang |
| author_facet | Yunbo Rao Menghan Zhang Zhanglin Cheng Junmin Xue Jiansu Pu Zairong Wang |
| author_sort | Yunbo Rao |
| collection | DOAJ |
| description | 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. |
| first_indexed | 2024-03-10T12:22:35Z |
| format | Article |
| id | doaj.art-cb49b1827e7343e4a905028e2af2eaa9 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T12:22:35Z |
| publishDate | 2021-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-cb49b1827e7343e4a905028e2af2eaa92023-11-21T15:21:58ZengMDPI AGSensors1424-82202021-04-01218273110.3390/s21082731Semantic Point Cloud Segmentation Using Fast Deep Neural Network and DCRFYunbo Rao0Menghan Zhang1Zhanglin Cheng2Junmin Xue3Jiansu Pu4Zairong Wang5School of Information and Software Engineering, University of Electronic Science and Technology of China, Chengdu 610054, ChinaSchool of Information and Software Engineering, University of Electronic Science and Technology of China, Chengdu 610054, ChinaShenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen 518055, ChinaSchool of Information and Software Engineering, University of Electronic Science and Technology of China, Chengdu 610054, ChinaSchool of Information and Software Engineering, University of Electronic Science and Technology of China, Chengdu 610054, ChinaSchool of Computer Science, Neijiang Normal University, Neijiang 641100, ChinaAccurate 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.https://www.mdpi.com/1424-8220/21/8/2731deep learning3D point clouddeep neural networksemantic segmentationDenseCRF |
| spellingShingle | Yunbo Rao Menghan Zhang Zhanglin Cheng Junmin Xue Jiansu Pu Zairong Wang Semantic Point Cloud Segmentation Using Fast Deep Neural Network and DCRF Sensors deep learning 3D point cloud deep neural network semantic segmentation DenseCRF |
| title | Semantic Point Cloud Segmentation Using Fast Deep Neural Network and DCRF |
| title_full | Semantic Point Cloud Segmentation Using Fast Deep Neural Network and DCRF |
| title_fullStr | Semantic Point Cloud Segmentation Using Fast Deep Neural Network and DCRF |
| title_full_unstemmed | Semantic Point Cloud Segmentation Using Fast Deep Neural Network and DCRF |
| title_short | Semantic Point Cloud Segmentation Using Fast Deep Neural Network and DCRF |
| title_sort | semantic point cloud segmentation using fast deep neural network and dcrf |
| topic | deep learning 3D point cloud deep neural network semantic segmentation DenseCRF |
| url | https://www.mdpi.com/1424-8220/21/8/2731 |
| work_keys_str_mv | AT yunborao semanticpointcloudsegmentationusingfastdeepneuralnetworkanddcrf AT menghanzhang semanticpointcloudsegmentationusingfastdeepneuralnetworkanddcrf AT zhanglincheng semanticpointcloudsegmentationusingfastdeepneuralnetworkanddcrf AT junminxue semanticpointcloudsegmentationusingfastdeepneuralnetworkanddcrf AT jiansupu semanticpointcloudsegmentationusingfastdeepneuralnetworkanddcrf AT zairongwang semanticpointcloudsegmentationusingfastdeepneuralnetworkanddcrf |