GeoNet: Deep Geodesic Networks for Point Cloud Analysis

Surface-based geodesic topology provides strong cues for object semantic analysis and geometric modeling. However, such connectivity information is lost in point clouds. Thus we introduce GeoNet, the first deep learning architecture trained to model the intrinsic structure of surfaces represented as...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: He, Tong, Huang, Haibin, Yi, Li, Zhou, Yuqian, Wu, Chihao, Wang, Jue, Soatto, Stefano
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator He, Tong
Huang, Haibin
Yi, Li
Zhou, Yuqian
Wu, Chihao
Wang, Jue
Soatto, Stefano
description Surface-based geodesic topology provides strong cues for object semantic analysis and geometric modeling. However, such connectivity information is lost in point clouds. Thus we introduce GeoNet, the first deep learning architecture trained to model the intrinsic structure of surfaces represented as point clouds. To demonstrate the applicability of learned geodesic-aware representations, we propose fusion schemes which use GeoNet in conjunction with other baseline or backbone networks, such as PU-Net and PointNet++, for down-stream point cloud analysis. Our method improves the state-of-the-art on multiple representative tasks that can benefit from understandings of the underlying surface topology, including point upsampling, normal estimation, mesh reconstruction and non-rigid shape classification.
doi_str_mv 10.48550/arxiv.1901.00680
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_1901_00680</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>1901_00680</sourcerecordid><originalsourceid>FETCH-LOGICAL-a670-2c93d1fb5a9e83cc845c910f157cb7d99801381456ac4219565c5f50e722d9423</originalsourceid><addsrcrecordid>eNotj8tOAjEYhbthQcAHYGVfYMa_l3_auiOjoAlRF-wnpZekcaSkxQtvD6Krk3MWX85HyIJBKzUi3Nnyk75aZoC1AJ2GKVHrkF_C8Z4-hHCgl-JDTY5epu9c3iuNudC3nPZH2o_509Pl3o6nmuqcTKIda7j5zxnZrh63_VOzeV0_98tNYzsFDXdGeBZ3aE3Qwjkt0RkGkaFyO-WN0cCEZhI76yRnBjt0GBGC4twbycWM3P5hr8-HQ0kftpyGX4PhaiDOjRw-tA</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>GeoNet: Deep Geodesic Networks for Point Cloud Analysis</title><source>arXiv.org</source><creator>He, Tong ; Huang, Haibin ; Yi, Li ; Zhou, Yuqian ; Wu, Chihao ; Wang, Jue ; Soatto, Stefano</creator><creatorcontrib>He, Tong ; Huang, Haibin ; Yi, Li ; Zhou, Yuqian ; Wu, Chihao ; Wang, Jue ; Soatto, Stefano</creatorcontrib><description>Surface-based geodesic topology provides strong cues for object semantic analysis and geometric modeling. However, such connectivity information is lost in point clouds. Thus we introduce GeoNet, the first deep learning architecture trained to model the intrinsic structure of surfaces represented as point clouds. To demonstrate the applicability of learned geodesic-aware representations, we propose fusion schemes which use GeoNet in conjunction with other baseline or backbone networks, such as PU-Net and PointNet++, for down-stream point cloud analysis. Our method improves the state-of-the-art on multiple representative tasks that can benefit from understandings of the underlying surface topology, including point upsampling, normal estimation, mesh reconstruction and non-rigid shape classification.</description><identifier>DOI: 10.48550/arxiv.1901.00680</identifier><language>eng</language><subject>Computer Science - Computer Vision and Pattern Recognition</subject><creationdate>2019-01</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,776,881</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/1901.00680$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.1901.00680$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>He, Tong</creatorcontrib><creatorcontrib>Huang, Haibin</creatorcontrib><creatorcontrib>Yi, Li</creatorcontrib><creatorcontrib>Zhou, Yuqian</creatorcontrib><creatorcontrib>Wu, Chihao</creatorcontrib><creatorcontrib>Wang, Jue</creatorcontrib><creatorcontrib>Soatto, Stefano</creatorcontrib><title>GeoNet: Deep Geodesic Networks for Point Cloud Analysis</title><description>Surface-based geodesic topology provides strong cues for object semantic analysis and geometric modeling. However, such connectivity information is lost in point clouds. Thus we introduce GeoNet, the first deep learning architecture trained to model the intrinsic structure of surfaces represented as point clouds. To demonstrate the applicability of learned geodesic-aware representations, we propose fusion schemes which use GeoNet in conjunction with other baseline or backbone networks, such as PU-Net and PointNet++, for down-stream point cloud analysis. Our method improves the state-of-the-art on multiple representative tasks that can benefit from understandings of the underlying surface topology, including point upsampling, normal estimation, mesh reconstruction and non-rigid shape classification.</description><subject>Computer Science - Computer Vision and Pattern Recognition</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2019</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotj8tOAjEYhbthQcAHYGVfYMa_l3_auiOjoAlRF-wnpZekcaSkxQtvD6Krk3MWX85HyIJBKzUi3Nnyk75aZoC1AJ2GKVHrkF_C8Z4-hHCgl-JDTY5epu9c3iuNudC3nPZH2o_509Pl3o6nmuqcTKIda7j5zxnZrh63_VOzeV0_98tNYzsFDXdGeBZ3aE3Qwjkt0RkGkaFyO-WN0cCEZhI76yRnBjt0GBGC4twbycWM3P5hr8-HQ0kftpyGX4PhaiDOjRw-tA</recordid><startdate>20190103</startdate><enddate>20190103</enddate><creator>He, Tong</creator><creator>Huang, Haibin</creator><creator>Yi, Li</creator><creator>Zhou, Yuqian</creator><creator>Wu, Chihao</creator><creator>Wang, Jue</creator><creator>Soatto, Stefano</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20190103</creationdate><title>GeoNet: Deep Geodesic Networks for Point Cloud Analysis</title><author>He, Tong ; Huang, Haibin ; Yi, Li ; Zhou, Yuqian ; Wu, Chihao ; Wang, Jue ; Soatto, Stefano</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a670-2c93d1fb5a9e83cc845c910f157cb7d99801381456ac4219565c5f50e722d9423</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2019</creationdate><topic>Computer Science - Computer Vision and Pattern Recognition</topic><toplevel>online_resources</toplevel><creatorcontrib>He, Tong</creatorcontrib><creatorcontrib>Huang, Haibin</creatorcontrib><creatorcontrib>Yi, Li</creatorcontrib><creatorcontrib>Zhou, Yuqian</creatorcontrib><creatorcontrib>Wu, Chihao</creatorcontrib><creatorcontrib>Wang, Jue</creatorcontrib><creatorcontrib>Soatto, Stefano</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>He, Tong</au><au>Huang, Haibin</au><au>Yi, Li</au><au>Zhou, Yuqian</au><au>Wu, Chihao</au><au>Wang, Jue</au><au>Soatto, Stefano</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>GeoNet: Deep Geodesic Networks for Point Cloud Analysis</atitle><date>2019-01-03</date><risdate>2019</risdate><abstract>Surface-based geodesic topology provides strong cues for object semantic analysis and geometric modeling. However, such connectivity information is lost in point clouds. Thus we introduce GeoNet, the first deep learning architecture trained to model the intrinsic structure of surfaces represented as point clouds. To demonstrate the applicability of learned geodesic-aware representations, we propose fusion schemes which use GeoNet in conjunction with other baseline or backbone networks, such as PU-Net and PointNet++, for down-stream point cloud analysis. Our method improves the state-of-the-art on multiple representative tasks that can benefit from understandings of the underlying surface topology, including point upsampling, normal estimation, mesh reconstruction and non-rigid shape classification.</abstract><doi>10.48550/arxiv.1901.00680</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.1901.00680
ispartof
issn
language eng
recordid cdi_arxiv_primary_1901_00680
source arXiv.org
subjects Computer Science - Computer Vision and Pattern Recognition
title GeoNet: Deep Geodesic Networks for Point Cloud Analysis
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-29T07%3A22%3A40IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=GeoNet:%20Deep%20Geodesic%20Networks%20for%20Point%20Cloud%20Analysis&rft.au=He,%20Tong&rft.date=2019-01-03&rft_id=info:doi/10.48550/arxiv.1901.00680&rft_dat=%3Carxiv_GOX%3E1901_00680%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true