Enhancing the performance of point cloud completion assisted with tactile information

Abstract Point cloud completion is a crucial task in the field of 3D imaging and is becoming a current research hotspot. However, traditional point cloud completion methods often fail to accurately restore the local details of objects due to the lack of detailed descriptions of the missing parts. Th...

Full description

Saved in:
Bibliographic Details
Main Authors: Ming Lu, Jian-Han Shi, Jian Li, Hao-Ran Shen, Huai-Yu Wang, Chi Zhang, Pu-Zheng Wang, Qin Wang
Format: Article
Language:English
Published: SpringerOpen 2024-12-01
Series:EURASIP Journal on Image and Video Processing
Subjects:
Online Access:https://doi.org/10.1186/s13640-024-00659-8
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849220618790109184
author Ming Lu
Jian-Han Shi
Jian Li
Hao-Ran Shen
Huai-Yu Wang
Chi Zhang
Pu-Zheng Wang
Qin Wang
author_facet Ming Lu
Jian-Han Shi
Jian Li
Hao-Ran Shen
Huai-Yu Wang
Chi Zhang
Pu-Zheng Wang
Qin Wang
author_sort Ming Lu
collection DOAJ
description Abstract Point cloud completion is a crucial task in the field of 3D imaging and is becoming a current research hotspot. However, traditional point cloud completion methods often fail to accurately restore the local details of objects due to the lack of detailed descriptions of the missing parts. This paper proposes a performance enhancement scheme for point cloud completion using tactile information. By employing a U-Net-based tactile feature extraction and fusion module, we enhance point cloud completion through the integration of tactile information. We validated our proposed network on the 3DVT dataset, and the experimental results demonstrate that our method effectively improves the performance of point cloud completion. Additionally, we conducted simulation experiments to mimic real-world scenarios where a robotic arm contacts objects, obtaining tactile point clouds using a mounted tactile simulator. These tactile point clouds were then used in related point cloud completion experiments, and the results were evaluated. The experimental results indicate that our proposed method significantly enhances the performance of point cloud completion and improves the local detail of the completed point clouds.
format Article
id doaj-art-1021d89ce5ab4a0eb9a1dc6b1bd356ff
institution Kabale University
issn 1687-5281
language English
publishDate 2024-12-01
publisher SpringerOpen
record_format Article
series EURASIP Journal on Image and Video Processing
spelling doaj-art-1021d89ce5ab4a0eb9a1dc6b1bd356ff2024-12-08T12:41:03ZengSpringerOpenEURASIP Journal on Image and Video Processing1687-52812024-12-012024111210.1186/s13640-024-00659-8Enhancing the performance of point cloud completion assisted with tactile informationMing Lu0Jian-Han Shi1Jian Li2Hao-Ran Shen3Huai-Yu Wang4Chi Zhang5Pu-Zheng Wang6Qin Wang7Institute of Quantum Information and Technology, Nanjing University of Posts and TelecommunicationsInstitute of Quantum Information and Technology, Nanjing University of Posts and TelecommunicationsInstitute of Quantum Information and Technology, Nanjing University of Posts and TelecommunicationsInstitute of Quantum Information and Technology, Nanjing University of Posts and TelecommunicationsInstitute of Quantum Information and Technology, Nanjing University of Posts and TelecommunicationsInstitute of Quantum Information and Technology, Nanjing University of Posts and TelecommunicationsInstitute of Quantum Information and Technology, Nanjing University of Posts and TelecommunicationsInstitute of Quantum Information and Technology, Nanjing University of Posts and TelecommunicationsAbstract Point cloud completion is a crucial task in the field of 3D imaging and is becoming a current research hotspot. However, traditional point cloud completion methods often fail to accurately restore the local details of objects due to the lack of detailed descriptions of the missing parts. This paper proposes a performance enhancement scheme for point cloud completion using tactile information. By employing a U-Net-based tactile feature extraction and fusion module, we enhance point cloud completion through the integration of tactile information. We validated our proposed network on the 3DVT dataset, and the experimental results demonstrate that our method effectively improves the performance of point cloud completion. Additionally, we conducted simulation experiments to mimic real-world scenarios where a robotic arm contacts objects, obtaining tactile point clouds using a mounted tactile simulator. These tactile point clouds were then used in related point cloud completion experiments, and the results were evaluated. The experimental results indicate that our proposed method significantly enhances the performance of point cloud completion and improves the local detail of the completed point clouds.https://doi.org/10.1186/s13640-024-00659-8Point cloudPoint cloud completeTacile assistedTacile fusion
spellingShingle Ming Lu
Jian-Han Shi
Jian Li
Hao-Ran Shen
Huai-Yu Wang
Chi Zhang
Pu-Zheng Wang
Qin Wang
Enhancing the performance of point cloud completion assisted with tactile information
EURASIP Journal on Image and Video Processing
Point cloud
Point cloud complete
Tacile assisted
Tacile fusion
title Enhancing the performance of point cloud completion assisted with tactile information
title_full Enhancing the performance of point cloud completion assisted with tactile information
title_fullStr Enhancing the performance of point cloud completion assisted with tactile information
title_full_unstemmed Enhancing the performance of point cloud completion assisted with tactile information
title_short Enhancing the performance of point cloud completion assisted with tactile information
title_sort enhancing the performance of point cloud completion assisted with tactile information
topic Point cloud
Point cloud complete
Tacile assisted
Tacile fusion
url https://doi.org/10.1186/s13640-024-00659-8
work_keys_str_mv AT minglu enhancingtheperformanceofpointcloudcompletionassistedwithtactileinformation
AT jianhanshi enhancingtheperformanceofpointcloudcompletionassistedwithtactileinformation
AT jianli enhancingtheperformanceofpointcloudcompletionassistedwithtactileinformation
AT haoranshen enhancingtheperformanceofpointcloudcompletionassistedwithtactileinformation
AT huaiyuwang enhancingtheperformanceofpointcloudcompletionassistedwithtactileinformation
AT chizhang enhancingtheperformanceofpointcloudcompletionassistedwithtactileinformation
AT puzhengwang enhancingtheperformanceofpointcloudcompletionassistedwithtactileinformation
AT qinwang enhancingtheperformanceofpointcloudcompletionassistedwithtactileinformation