Revisiting PatchMatch Multi-View Stereo for Urban 3D Reconstruction

Revisiting PatchMatch Multi-View Stereo for Urban 3D Reconstruction

In this paper, a complete pipeline for image-based 3D reconstruction of urban scenarios is proposed, based on PatchMatch Multi-View Stereo (MVS). Input images are firstly fed into an off-the-shelf visual SLAM system to extract camera poses and sparse keypoints, which are used to initialize PatchMatc...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Orsingher, Marco, Zani, Paolo, Medici, Paolo, Bertozzi, Massimo
Format: Artikel
Sprache:eng
Schlagworte:
Computer Science - Computer Vision and Pattern Recognition
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 Orsingher, Marco
Zani, Paolo
Medici, Paolo
Bertozzi, Massimo
description In this paper, a complete pipeline for image-based 3D reconstruction of urban scenarios is proposed, based on PatchMatch Multi-View Stereo (MVS). Input images are firstly fed into an off-the-shelf visual SLAM system to extract camera poses and sparse keypoints, which are used to initialize PatchMatch optimization. Then, pixelwise depths and normals are iteratively computed in a multi-scale framework with a novel depth-normal consistency loss term and a global refinement algorithm to balance the inherently local nature of PatchMatch. Finally, a large-scale point cloud is generated by back-projecting multi-view consistent estimates in 3D. The proposed approach is carefully evaluated against both classical MVS algorithms and monocular depth networks on the KITTI dataset, showing state of the art performances.
doi_str_mv 10.48550/arxiv.2207.08439
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2207_08439</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2207_08439</sourcerecordid><originalsourceid>FETCH-LOGICAL-a679-47ca1520fb631c451cbf9075d17bfe1ddf0e124c2f0cc02178603993450707243</originalsourceid><addsrcrecordid>eNotz8lOwzAUhWFvWKDCA7CqXyDheorjZRVGqRWo0zZybuxiqSTIcQfevuqwOf_uSB8hTwxyWSoFzzYewz7nHHQOpRTmnlRztw9DSKHb0G-b8Gd2HjrbbVPI1sEd6CK56Hrq-0hXsbEdFS907rDvhhR3mELfPZA7b7eDe7x1RJZvr8vqI5t-vX9Wk2lmC20yqdEyxcE3hWAoFcPGG9CqZbrxjrWtB8e4RO4BETjTZQHCGCEVaNBcihEZX28vivovhl8b_-uzpr5oxAm4ZkO5</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Revisiting PatchMatch Multi-View Stereo for Urban 3D Reconstruction</title><source>arXiv.org</source><creator>Orsingher, Marco ; Zani, Paolo ; Medici, Paolo ; Bertozzi, Massimo</creator><creatorcontrib>Orsingher, Marco ; Zani, Paolo ; Medici, Paolo ; Bertozzi, Massimo</creatorcontrib><description>In this paper, a complete pipeline for image-based 3D reconstruction of urban scenarios is proposed, based on PatchMatch Multi-View Stereo (MVS). Input images are firstly fed into an off-the-shelf visual SLAM system to extract camera poses and sparse keypoints, which are used to initialize PatchMatch optimization. Then, pixelwise depths and normals are iteratively computed in a multi-scale framework with a novel depth-normal consistency loss term and a global refinement algorithm to balance the inherently local nature of PatchMatch. Finally, a large-scale point cloud is generated by back-projecting multi-view consistent estimates in 3D. The proposed approach is carefully evaluated against both classical MVS algorithms and monocular depth networks on the KITTI dataset, showing state of the art performances.</description><identifier>DOI: 10.48550/arxiv.2207.08439</identifier><language>eng</language><subject>Computer Science - Computer Vision and Pattern Recognition</subject><creationdate>2022-07</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>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,780,885</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2207.08439$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2207.08439$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Orsingher, Marco</creatorcontrib><creatorcontrib>Zani, Paolo</creatorcontrib><creatorcontrib>Medici, Paolo</creatorcontrib><creatorcontrib>Bertozzi, Massimo</creatorcontrib><title>Revisiting PatchMatch Multi-View Stereo for Urban 3D Reconstruction</title><description>In this paper, a complete pipeline for image-based 3D reconstruction of urban scenarios is proposed, based on PatchMatch Multi-View Stereo (MVS). Input images are firstly fed into an off-the-shelf visual SLAM system to extract camera poses and sparse keypoints, which are used to initialize PatchMatch optimization. Then, pixelwise depths and normals are iteratively computed in a multi-scale framework with a novel depth-normal consistency loss term and a global refinement algorithm to balance the inherently local nature of PatchMatch. Finally, a large-scale point cloud is generated by back-projecting multi-view consistent estimates in 3D. The proposed approach is carefully evaluated against both classical MVS algorithms and monocular depth networks on the KITTI dataset, showing state of the art performances.</description><subject>Computer Science - Computer Vision and Pattern Recognition</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2022</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotz8lOwzAUhWFvWKDCA7CqXyDheorjZRVGqRWo0zZybuxiqSTIcQfevuqwOf_uSB8hTwxyWSoFzzYewz7nHHQOpRTmnlRztw9DSKHb0G-b8Gd2HjrbbVPI1sEd6CK56Hrq-0hXsbEdFS907rDvhhR3mELfPZA7b7eDe7x1RJZvr8vqI5t-vX9Wk2lmC20yqdEyxcE3hWAoFcPGG9CqZbrxjrWtB8e4RO4BETjTZQHCGCEVaNBcihEZX28vivovhl8b_-uzpr5oxAm4ZkO5</recordid><startdate>20220718</startdate><enddate>20220718</enddate><creator>Orsingher, Marco</creator><creator>Zani, Paolo</creator><creator>Medici, Paolo</creator><creator>Bertozzi, Massimo</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20220718</creationdate><title>Revisiting PatchMatch Multi-View Stereo for Urban 3D Reconstruction</title><author>Orsingher, Marco ; Zani, Paolo ; Medici, Paolo ; Bertozzi, Massimo</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a679-47ca1520fb631c451cbf9075d17bfe1ddf0e124c2f0cc02178603993450707243</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2022</creationdate><topic>Computer Science - Computer Vision and Pattern Recognition</topic><toplevel>online_resources</toplevel><creatorcontrib>Orsingher, Marco</creatorcontrib><creatorcontrib>Zani, Paolo</creatorcontrib><creatorcontrib>Medici, Paolo</creatorcontrib><creatorcontrib>Bertozzi, Massimo</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Orsingher, Marco</au><au>Zani, Paolo</au><au>Medici, Paolo</au><au>Bertozzi, Massimo</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Revisiting PatchMatch Multi-View Stereo for Urban 3D Reconstruction</atitle><date>2022-07-18</date><risdate>2022</risdate><abstract>In this paper, a complete pipeline for image-based 3D reconstruction of urban scenarios is proposed, based on PatchMatch Multi-View Stereo (MVS). Input images are firstly fed into an off-the-shelf visual SLAM system to extract camera poses and sparse keypoints, which are used to initialize PatchMatch optimization. Then, pixelwise depths and normals are iteratively computed in a multi-scale framework with a novel depth-normal consistency loss term and a global refinement algorithm to balance the inherently local nature of PatchMatch. Finally, a large-scale point cloud is generated by back-projecting multi-view consistent estimates in 3D. The proposed approach is carefully evaluated against both classical MVS algorithms and monocular depth networks on the KITTI dataset, showing state of the art performances.</abstract><doi>10.48550/arxiv.2207.08439</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2207.08439
ispartof
issn
language eng
recordid cdi_arxiv_primary_2207_08439
source arXiv.org
subjects Computer Science - Computer Vision and Pattern Recognition
title Revisiting PatchMatch Multi-View Stereo for Urban 3D Reconstruction
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-10T15%3A23%3A45IST&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=Revisiting%20PatchMatch%20Multi-View%20Stereo%20for%20Urban%203D%20Reconstruction&rft.au=Orsingher,%20Marco&rft.date=2022-07-18&rft_id=info:doi/10.48550/arxiv.2207.08439&rft_dat=%3Carxiv_GOX%3E2207_08439%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