DQRNet: Dynamic Quality Refinement Network for 3D Reconstruction from a Single Depth View

With the widespread adoption of 3D scanning technology, depth view-driven 3D reconstruction has become crucial for applications such as SLAM, virtual reality, and autonomous vehicles. However, due to the effects of self-occlusion and environmental occlusion, obtaining complete and error-free 3D shap...

Full description

Saved in:
Bibliographic Details
Main Authors: Caixia Liu, Minhong Zhu, Haisheng Li, Xiulan Wei, Jiulin Liang, Qianwen Yao
Format: Article
Language:English
Published: MDPI AG 2025-02-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/25/5/1503
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850053052184133632
author Caixia Liu
Minhong Zhu
Haisheng Li
Xiulan Wei
Jiulin Liang
Qianwen Yao
author_facet Caixia Liu
Minhong Zhu
Haisheng Li
Xiulan Wei
Jiulin Liang
Qianwen Yao
author_sort Caixia Liu
collection DOAJ
description With the widespread adoption of 3D scanning technology, depth view-driven 3D reconstruction has become crucial for applications such as SLAM, virtual reality, and autonomous vehicles. However, due to the effects of self-occlusion and environmental occlusion, obtaining complete and error-free 3D shapes directly from 3D scans remains challenging, as previous reconstruction methods tend to lose details. To this end, we propose Dynamic Quality Refinement Network (DQRNet) for reconstructing complete and accurate 3D shape from a single depth view. DQRNet introduces a dynamic encoder–decoder and a detail quality refiner to generate high-resolution 3D shapes, where the former designs a dynamic latent extractor to adaptively select important parts of an object and the latter designs global and local point refiners to enhance the reconstruction quality. Experimental results show that DQRNet is able to focus on capturing the details at boundaries and key areas on ShapeNet dataset, thereby achieving better accuracy and robustness than SOTA methods.
format Article
id doaj-art-f2f7a8b9b8e4493dbd3cef28d9d39dc9
institution DOAJ
issn 1424-8220
language English
publishDate 2025-02-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj-art-f2f7a8b9b8e4493dbd3cef28d9d39dc92025-08-20T02:52:38ZengMDPI AGSensors1424-82202025-02-01255150310.3390/s25051503DQRNet: Dynamic Quality Refinement Network for 3D Reconstruction from a Single Depth ViewCaixia Liu0Minhong Zhu1Haisheng Li2Xiulan Wei3Jiulin Liang4Qianwen Yao5Beijing Key Laboratory of Big Data Technology for Food Safety, School of Computer and Artificial Intelligence, Beijing Technology and Business University, No. 33, Fucheng Road, Haidian District, Beijing 100048, ChinaBeijing Key Laboratory of Big Data Technology for Food Safety, School of Computer and Artificial Intelligence, Beijing Technology and Business University, No. 33, Fucheng Road, Haidian District, Beijing 100048, ChinaBeijing Key Laboratory of Big Data Technology for Food Safety, School of Computer and Artificial Intelligence, Beijing Technology and Business University, No. 33, Fucheng Road, Haidian District, Beijing 100048, ChinaSchool of Logistics, Beijing Wuzi University, No. 321, Fuhe Street, Tongzhou District, Beijing 101149, ChinaBeijing Key Laboratory of Big Data Technology for Food Safety, School of Computer and Artificial Intelligence, Beijing Technology and Business University, No. 33, Fucheng Road, Haidian District, Beijing 100048, ChinaBeijing Key Laboratory of Big Data Technology for Food Safety, School of Computer and Artificial Intelligence, Beijing Technology and Business University, No. 33, Fucheng Road, Haidian District, Beijing 100048, ChinaWith the widespread adoption of 3D scanning technology, depth view-driven 3D reconstruction has become crucial for applications such as SLAM, virtual reality, and autonomous vehicles. However, due to the effects of self-occlusion and environmental occlusion, obtaining complete and error-free 3D shapes directly from 3D scans remains challenging, as previous reconstruction methods tend to lose details. To this end, we propose Dynamic Quality Refinement Network (DQRNet) for reconstructing complete and accurate 3D shape from a single depth view. DQRNet introduces a dynamic encoder–decoder and a detail quality refiner to generate high-resolution 3D shapes, where the former designs a dynamic latent extractor to adaptively select important parts of an object and the latter designs global and local point refiners to enhance the reconstruction quality. Experimental results show that DQRNet is able to focus on capturing the details at boundaries and key areas on ShapeNet dataset, thereby achieving better accuracy and robustness than SOTA methods.https://www.mdpi.com/1424-8220/25/5/15033D shape completiondynamic encoder–decoderglobal and local point refinerssingle depth view
spellingShingle Caixia Liu
Minhong Zhu
Haisheng Li
Xiulan Wei
Jiulin Liang
Qianwen Yao
DQRNet: Dynamic Quality Refinement Network for 3D Reconstruction from a Single Depth View
Sensors
3D shape completion
dynamic encoder–decoder
global and local point refiners
single depth view
title DQRNet: Dynamic Quality Refinement Network for 3D Reconstruction from a Single Depth View
title_full DQRNet: Dynamic Quality Refinement Network for 3D Reconstruction from a Single Depth View
title_fullStr DQRNet: Dynamic Quality Refinement Network for 3D Reconstruction from a Single Depth View
title_full_unstemmed DQRNet: Dynamic Quality Refinement Network for 3D Reconstruction from a Single Depth View
title_short DQRNet: Dynamic Quality Refinement Network for 3D Reconstruction from a Single Depth View
title_sort dqrnet dynamic quality refinement network for 3d reconstruction from a single depth view
topic 3D shape completion
dynamic encoder–decoder
global and local point refiners
single depth view
url https://www.mdpi.com/1424-8220/25/5/1503
work_keys_str_mv AT caixialiu dqrnetdynamicqualityrefinementnetworkfor3dreconstructionfromasingledepthview
AT minhongzhu dqrnetdynamicqualityrefinementnetworkfor3dreconstructionfromasingledepthview
AT haishengli dqrnetdynamicqualityrefinementnetworkfor3dreconstructionfromasingledepthview
AT xiulanwei dqrnetdynamicqualityrefinementnetworkfor3dreconstructionfromasingledepthview
AT jiulinliang dqrnetdynamicqualityrefinementnetworkfor3dreconstructionfromasingledepthview
AT qianwenyao dqrnetdynamicqualityrefinementnetworkfor3dreconstructionfromasingledepthview