COSMU: Complete 3D human shape from monocular unconstrained images

We present a novel framework to reconstruct complete 3D human shapes from a given target image by leveraging monocular unconstrained images. The objective of this work is to reproduce high-quality details in regions of the reconstructed human body that are not visible in the input target. The propos...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Pesavento, Marco, Volino, Marco, Hilton, Adrian
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 Pesavento, Marco
Volino, Marco
Hilton, Adrian
description We present a novel framework to reconstruct complete 3D human shapes from a given target image by leveraging monocular unconstrained images. The objective of this work is to reproduce high-quality details in regions of the reconstructed human body that are not visible in the input target. The proposed methodology addresses the limitations of existing approaches for reconstructing 3D human shapes from a single image, which cannot reproduce shape details in occluded body regions. The missing information of the monocular input can be recovered by using multiple views captured from multiple cameras. However, multi-view reconstruction methods necessitate accurately calibrated and registered images, which can be challenging to obtain in real-world scenarios. Given a target RGB image and a collection of multiple uncalibrated and unregistered images of the same individual, acquired using a single camera, we propose a novel framework to generate complete 3D human shapes. We introduce a novel module to generate 2D multi-view normal maps of the person registered with the target input image. The module consists of body part-based reference selection and body part-based registration. The generated 2D normal maps are then processed by a multi-view attention-based neural implicit model that estimates an implicit representation of the 3D shape, ensuring the reproduction of details in both observed and occluded regions. Extensive experiments demonstrate that the proposed approach estimates higher quality details in the non-visible regions of the 3D clothed human shapes compared to related methods, without using parametric models.
doi_str_mv 10.48550/arxiv.2407.10586
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2407_10586</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2407_10586</sourcerecordid><originalsourceid>FETCH-arxiv_primary_2407_105863</originalsourceid><addsrcrecordid>eNpjYJA0NNAzsTA1NdBPLKrILNMzMjEw1zM0MLUw42RwcvYP9g21UnDOzy3ISS1JVTB2UcgozU3MUyjOSCxIVUgrys9VyM3Py08uzUksUijNS87PKy4pSszMS01RyMxNTE8t5mFgTUvMKU7lhdLcDPJuriHOHrpg2-ILioDKiirjQbbGg201JqwCAAwGNus</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>COSMU: Complete 3D human shape from monocular unconstrained images</title><source>arXiv.org</source><creator>Pesavento, Marco ; Volino, Marco ; Hilton, Adrian</creator><creatorcontrib>Pesavento, Marco ; Volino, Marco ; Hilton, Adrian</creatorcontrib><description>We present a novel framework to reconstruct complete 3D human shapes from a given target image by leveraging monocular unconstrained images. The objective of this work is to reproduce high-quality details in regions of the reconstructed human body that are not visible in the input target. The proposed methodology addresses the limitations of existing approaches for reconstructing 3D human shapes from a single image, which cannot reproduce shape details in occluded body regions. The missing information of the monocular input can be recovered by using multiple views captured from multiple cameras. However, multi-view reconstruction methods necessitate accurately calibrated and registered images, which can be challenging to obtain in real-world scenarios. Given a target RGB image and a collection of multiple uncalibrated and unregistered images of the same individual, acquired using a single camera, we propose a novel framework to generate complete 3D human shapes. We introduce a novel module to generate 2D multi-view normal maps of the person registered with the target input image. The module consists of body part-based reference selection and body part-based registration. The generated 2D normal maps are then processed by a multi-view attention-based neural implicit model that estimates an implicit representation of the 3D shape, ensuring the reproduction of details in both observed and occluded regions. Extensive experiments demonstrate that the proposed approach estimates higher quality details in the non-visible regions of the 3D clothed human shapes compared to related methods, without using parametric models.</description><identifier>DOI: 10.48550/arxiv.2407.10586</identifier><language>eng</language><subject>Computer Science - Computer Vision and Pattern Recognition</subject><creationdate>2024-07</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>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,776,881</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2407.10586$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2407.10586$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Pesavento, Marco</creatorcontrib><creatorcontrib>Volino, Marco</creatorcontrib><creatorcontrib>Hilton, Adrian</creatorcontrib><title>COSMU: Complete 3D human shape from monocular unconstrained images</title><description>We present a novel framework to reconstruct complete 3D human shapes from a given target image by leveraging monocular unconstrained images. The objective of this work is to reproduce high-quality details in regions of the reconstructed human body that are not visible in the input target. The proposed methodology addresses the limitations of existing approaches for reconstructing 3D human shapes from a single image, which cannot reproduce shape details in occluded body regions. The missing information of the monocular input can be recovered by using multiple views captured from multiple cameras. However, multi-view reconstruction methods necessitate accurately calibrated and registered images, which can be challenging to obtain in real-world scenarios. Given a target RGB image and a collection of multiple uncalibrated and unregistered images of the same individual, acquired using a single camera, we propose a novel framework to generate complete 3D human shapes. We introduce a novel module to generate 2D multi-view normal maps of the person registered with the target input image. The module consists of body part-based reference selection and body part-based registration. The generated 2D normal maps are then processed by a multi-view attention-based neural implicit model that estimates an implicit representation of the 3D shape, ensuring the reproduction of details in both observed and occluded regions. Extensive experiments demonstrate that the proposed approach estimates higher quality details in the non-visible regions of the 3D clothed human shapes compared to related methods, without using parametric models.</description><subject>Computer Science - Computer Vision and Pattern Recognition</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNpjYJA0NNAzsTA1NdBPLKrILNMzMjEw1zM0MLUw42RwcvYP9g21UnDOzy3ISS1JVTB2UcgozU3MUyjOSCxIVUgrys9VyM3Py08uzUksUijNS87PKy4pSszMS01RyMxNTE8t5mFgTUvMKU7lhdLcDPJuriHOHrpg2-ILioDKiirjQbbGg201JqwCAAwGNus</recordid><startdate>20240715</startdate><enddate>20240715</enddate><creator>Pesavento, Marco</creator><creator>Volino, Marco</creator><creator>Hilton, Adrian</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20240715</creationdate><title>COSMU: Complete 3D human shape from monocular unconstrained images</title><author>Pesavento, Marco ; Volino, Marco ; Hilton, Adrian</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-arxiv_primary_2407_105863</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Computer Science - Computer Vision and Pattern Recognition</topic><toplevel>online_resources</toplevel><creatorcontrib>Pesavento, Marco</creatorcontrib><creatorcontrib>Volino, Marco</creatorcontrib><creatorcontrib>Hilton, Adrian</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Pesavento, Marco</au><au>Volino, Marco</au><au>Hilton, Adrian</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>COSMU: Complete 3D human shape from monocular unconstrained images</atitle><date>2024-07-15</date><risdate>2024</risdate><abstract>We present a novel framework to reconstruct complete 3D human shapes from a given target image by leveraging monocular unconstrained images. The objective of this work is to reproduce high-quality details in regions of the reconstructed human body that are not visible in the input target. The proposed methodology addresses the limitations of existing approaches for reconstructing 3D human shapes from a single image, which cannot reproduce shape details in occluded body regions. The missing information of the monocular input can be recovered by using multiple views captured from multiple cameras. However, multi-view reconstruction methods necessitate accurately calibrated and registered images, which can be challenging to obtain in real-world scenarios. Given a target RGB image and a collection of multiple uncalibrated and unregistered images of the same individual, acquired using a single camera, we propose a novel framework to generate complete 3D human shapes. We introduce a novel module to generate 2D multi-view normal maps of the person registered with the target input image. The module consists of body part-based reference selection and body part-based registration. The generated 2D normal maps are then processed by a multi-view attention-based neural implicit model that estimates an implicit representation of the 3D shape, ensuring the reproduction of details in both observed and occluded regions. Extensive experiments demonstrate that the proposed approach estimates higher quality details in the non-visible regions of the 3D clothed human shapes compared to related methods, without using parametric models.</abstract><doi>10.48550/arxiv.2407.10586</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2407.10586
ispartof
issn
language eng
recordid cdi_arxiv_primary_2407_10586
source arXiv.org
subjects Computer Science - Computer Vision and Pattern Recognition
title COSMU: Complete 3D human shape from monocular unconstrained images
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-25T07%3A28%3A10IST&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=COSMU:%20Complete%203D%20human%20shape%20from%20monocular%20unconstrained%20images&rft.au=Pesavento,%20Marco&rft.date=2024-07-15&rft_id=info:doi/10.48550/arxiv.2407.10586&rft_dat=%3Carxiv_GOX%3E2407_10586%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