SPAC-Net: Rethinking Point Cloud Completion with Structural Prior

Point cloud completion aims to infer a complete shape from its partial observation. Many approaches utilize a pure encoderdecoder paradigm in which complete shape can be directly predicted by shape priors learned from partial scans, however, these methods suffer from the loss of details inevitably d...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:arXiv.org 2024-11
Hauptverfasser: Wu, Zizhao, Shi, Jian, Deng, Xuan, Zhang, Cheng, Yang, Genfu, Zeng, Ming, Wang, Yunhai
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title arXiv.org
container_volume
creator Wu, Zizhao
Shi, Jian
Deng, Xuan
Zhang, Cheng
Yang, Genfu
Zeng, Ming
Wang, Yunhai
description Point cloud completion aims to infer a complete shape from its partial observation. Many approaches utilize a pure encoderdecoder paradigm in which complete shape can be directly predicted by shape priors learned from partial scans, however, these methods suffer from the loss of details inevitably due to the feature abstraction issues. In this paper, we propose a novel framework,termed SPAC-Net, that aims to rethink the completion task under the guidance of a new structural prior, we call it interface. Specifically, our method first investigates Marginal Detector (MAD) module to localize the interface, defined as the intersection between the known observation and the missing parts. Based on the interface, our method predicts the coarse shape by learning the displacement from the points in interface move to their corresponding position in missing parts. Furthermore, we devise an additional Structure Supplement(SSP) module before the upsampling stage to enhance the structural details of the coarse shape, enabling the upsampling module to focus more on the upsampling task. Extensive experiments have been conducted on several challenging benchmarks, and the results demonstrate that our method outperforms existing state-of-the-art approaches.
format Article
fullrecord <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_3132694490</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>3132694490</sourcerecordid><originalsourceid>FETCH-proquest_journals_31326944903</originalsourceid><addsrcrecordid>eNqNykELgjAYgOERBEn5HwadhbmpZTcZRaeQ9C5SK2drn23f6O_XoR_Q6T0874xEXIg02WacL0js_cgY48WG57mISNXUlUxOCnf0rHDQ9qHtndagLVJpIFyphOdkFGqw9K1xoA26cMHgekNrp8GtyPzWG6_iX5dkfdi38phMDl5BeexGCM5-qROp4EWZZSUT_10fFZo43A</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>3132694490</pqid></control><display><type>article</type><title>SPAC-Net: Rethinking Point Cloud Completion with Structural Prior</title><source>Free E- Journals</source><creator>Wu, Zizhao ; Shi, Jian ; Deng, Xuan ; Zhang, Cheng ; Yang, Genfu ; Zeng, Ming ; Wang, Yunhai</creator><creatorcontrib>Wu, Zizhao ; Shi, Jian ; Deng, Xuan ; Zhang, Cheng ; Yang, Genfu ; Zeng, Ming ; Wang, Yunhai</creatorcontrib><description>Point cloud completion aims to infer a complete shape from its partial observation. Many approaches utilize a pure encoderdecoder paradigm in which complete shape can be directly predicted by shape priors learned from partial scans, however, these methods suffer from the loss of details inevitably due to the feature abstraction issues. In this paper, we propose a novel framework,termed SPAC-Net, that aims to rethink the completion task under the guidance of a new structural prior, we call it interface. Specifically, our method first investigates Marginal Detector (MAD) module to localize the interface, defined as the intersection between the known observation and the missing parts. Based on the interface, our method predicts the coarse shape by learning the displacement from the points in interface move to their corresponding position in missing parts. Furthermore, we devise an additional Structure Supplement(SSP) module before the upsampling stage to enhance the structural details of the coarse shape, enabling the upsampling module to focus more on the upsampling task. Extensive experiments have been conducted on several challenging benchmarks, and the results demonstrate that our method outperforms existing state-of-the-art approaches.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Modules</subject><ispartof>arXiv.org, 2024-11</ispartof><rights>2024. This work is published under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>780,784</link.rule.ids></links><search><creatorcontrib>Wu, Zizhao</creatorcontrib><creatorcontrib>Shi, Jian</creatorcontrib><creatorcontrib>Deng, Xuan</creatorcontrib><creatorcontrib>Zhang, Cheng</creatorcontrib><creatorcontrib>Yang, Genfu</creatorcontrib><creatorcontrib>Zeng, Ming</creatorcontrib><creatorcontrib>Wang, Yunhai</creatorcontrib><title>SPAC-Net: Rethinking Point Cloud Completion with Structural Prior</title><title>arXiv.org</title><description>Point cloud completion aims to infer a complete shape from its partial observation. Many approaches utilize a pure encoderdecoder paradigm in which complete shape can be directly predicted by shape priors learned from partial scans, however, these methods suffer from the loss of details inevitably due to the feature abstraction issues. In this paper, we propose a novel framework,termed SPAC-Net, that aims to rethink the completion task under the guidance of a new structural prior, we call it interface. Specifically, our method first investigates Marginal Detector (MAD) module to localize the interface, defined as the intersection between the known observation and the missing parts. Based on the interface, our method predicts the coarse shape by learning the displacement from the points in interface move to their corresponding position in missing parts. Furthermore, we devise an additional Structure Supplement(SSP) module before the upsampling stage to enhance the structural details of the coarse shape, enabling the upsampling module to focus more on the upsampling task. Extensive experiments have been conducted on several challenging benchmarks, and the results demonstrate that our method outperforms existing state-of-the-art approaches.</description><subject>Modules</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><sourceid>ABUWG</sourceid><sourceid>AFKRA</sourceid><sourceid>AZQEC</sourceid><sourceid>BENPR</sourceid><sourceid>CCPQU</sourceid><sourceid>DWQXO</sourceid><recordid>eNqNykELgjAYgOERBEn5HwadhbmpZTcZRaeQ9C5SK2drn23f6O_XoR_Q6T0874xEXIg02WacL0js_cgY48WG57mISNXUlUxOCnf0rHDQ9qHtndagLVJpIFyphOdkFGqw9K1xoA26cMHgekNrp8GtyPzWG6_iX5dkfdi38phMDl5BeexGCM5-qROp4EWZZSUT_10fFZo43A</recordid><startdate>20241122</startdate><enddate>20241122</enddate><creator>Wu, Zizhao</creator><creator>Shi, Jian</creator><creator>Deng, Xuan</creator><creator>Zhang, Cheng</creator><creator>Yang, Genfu</creator><creator>Zeng, Ming</creator><creator>Wang, Yunhai</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope></search><sort><creationdate>20241122</creationdate><title>SPAC-Net: Rethinking Point Cloud Completion with Structural Prior</title><author>Wu, Zizhao ; Shi, Jian ; Deng, Xuan ; Zhang, Cheng ; Yang, Genfu ; Zeng, Ming ; Wang, Yunhai</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_31326944903</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Modules</topic><toplevel>online_resources</toplevel><creatorcontrib>Wu, Zizhao</creatorcontrib><creatorcontrib>Shi, Jian</creatorcontrib><creatorcontrib>Deng, Xuan</creatorcontrib><creatorcontrib>Zhang, Cheng</creatorcontrib><creatorcontrib>Yang, Genfu</creatorcontrib><creatorcontrib>Zeng, Ming</creatorcontrib><creatorcontrib>Wang, Yunhai</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science &amp; Engineering Collection</collection><collection>ProQuest Central (Alumni Edition)</collection><collection>ProQuest Central UK/Ireland</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>Publicly Available Content Database</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering Collection</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Wu, Zizhao</au><au>Shi, Jian</au><au>Deng, Xuan</au><au>Zhang, Cheng</au><au>Yang, Genfu</au><au>Zeng, Ming</au><au>Wang, Yunhai</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>SPAC-Net: Rethinking Point Cloud Completion with Structural Prior</atitle><jtitle>arXiv.org</jtitle><date>2024-11-22</date><risdate>2024</risdate><eissn>2331-8422</eissn><abstract>Point cloud completion aims to infer a complete shape from its partial observation. Many approaches utilize a pure encoderdecoder paradigm in which complete shape can be directly predicted by shape priors learned from partial scans, however, these methods suffer from the loss of details inevitably due to the feature abstraction issues. In this paper, we propose a novel framework,termed SPAC-Net, that aims to rethink the completion task under the guidance of a new structural prior, we call it interface. Specifically, our method first investigates Marginal Detector (MAD) module to localize the interface, defined as the intersection between the known observation and the missing parts. Based on the interface, our method predicts the coarse shape by learning the displacement from the points in interface move to their corresponding position in missing parts. Furthermore, we devise an additional Structure Supplement(SSP) module before the upsampling stage to enhance the structural details of the coarse shape, enabling the upsampling module to focus more on the upsampling task. Extensive experiments have been conducted on several challenging benchmarks, and the results demonstrate that our method outperforms existing state-of-the-art approaches.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier EISSN: 2331-8422
ispartof arXiv.org, 2024-11
issn 2331-8422
language eng
recordid cdi_proquest_journals_3132694490
source Free E- Journals
subjects Modules
title SPAC-Net: Rethinking Point Cloud Completion with Structural Prior
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-24T20%3A01%3A38IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=document&rft.atitle=SPAC-Net:%20Rethinking%20Point%20Cloud%20Completion%20with%20Structural%20Prior&rft.jtitle=arXiv.org&rft.au=Wu,%20Zizhao&rft.date=2024-11-22&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E3132694490%3C/proquest%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=3132694490&rft_id=info:pmid/&rfr_iscdi=true