OmniRe: Omni Urban Scene Reconstruction

We introduce OmniRe, a holistic approach for efficiently reconstructing high-fidelity dynamic urban scenes from on-device logs. Recent methods for modeling driving sequences using neural radiance fields or Gaussian Splatting have demonstrated the potential of reconstructing challenging dynamic scene...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Chen, Ziyu, Yang, Jiawei, Huang, Jiahui, de Lutio, Riccardo, Esturo, Janick Martinez, Ivanovic, Boris, Litany, Or, Gojcic, Zan, Fidler, Sanja, Pavone, Marco, Song, Li, Wang, Yue
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 Chen, Ziyu
Yang, Jiawei
Huang, Jiahui
de Lutio, Riccardo
Esturo, Janick Martinez
Ivanovic, Boris
Litany, Or
Gojcic, Zan
Fidler, Sanja
Pavone, Marco
Song, Li
Wang, Yue
description We introduce OmniRe, a holistic approach for efficiently reconstructing high-fidelity dynamic urban scenes from on-device logs. Recent methods for modeling driving sequences using neural radiance fields or Gaussian Splatting have demonstrated the potential of reconstructing challenging dynamic scenes, but often overlook pedestrians and other non-vehicle dynamic actors, hindering a complete pipeline for dynamic urban scene reconstruction. To that end, we propose a comprehensive 3DGS framework for driving scenes, named OmniRe, that allows for accurate, full-length reconstruction of diverse dynamic objects in a driving log. OmniRe builds dynamic neural scene graphs based on Gaussian representations and constructs multiple local canonical spaces that model various dynamic actors, including vehicles, pedestrians, and cyclists, among many others. This capability is unmatched by existing methods. OmniRe allows us to holistically reconstruct different objects present in the scene, subsequently enabling the simulation of reconstructed scenarios with all actors participating in real-time (~60Hz). Extensive evaluations on the Waymo dataset show that our approach outperforms prior state-of-the-art methods quantitatively and qualitatively by a large margin. We believe our work fills a critical gap in driving reconstruction.
doi_str_mv 10.48550/arxiv.2408.16760
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2408_16760</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2408_16760</sourcerecordid><originalsourceid>FETCH-arxiv_primary_2408_167603</originalsourceid><addsrcrecordid>eNpjYJA0NNAzsTA1NdBPLKrILNMzMjGw0DM0Mzcz4GRQ98_NywxKtVIA0QqhRUmJeQrByal5qQpBqcn5ecUlRaXJJZn5eTwMrGmJOcWpvFCam0HezTXE2UMXbGR8QVFmbmJRZTzI6Hiw0caEVQAANCYtIA</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>OmniRe: Omni Urban Scene Reconstruction</title><source>arXiv.org</source><creator>Chen, Ziyu ; Yang, Jiawei ; Huang, Jiahui ; de Lutio, Riccardo ; Esturo, Janick Martinez ; Ivanovic, Boris ; Litany, Or ; Gojcic, Zan ; Fidler, Sanja ; Pavone, Marco ; Song, Li ; Wang, Yue</creator><creatorcontrib>Chen, Ziyu ; Yang, Jiawei ; Huang, Jiahui ; de Lutio, Riccardo ; Esturo, Janick Martinez ; Ivanovic, Boris ; Litany, Or ; Gojcic, Zan ; Fidler, Sanja ; Pavone, Marco ; Song, Li ; Wang, Yue</creatorcontrib><description>We introduce OmniRe, a holistic approach for efficiently reconstructing high-fidelity dynamic urban scenes from on-device logs. Recent methods for modeling driving sequences using neural radiance fields or Gaussian Splatting have demonstrated the potential of reconstructing challenging dynamic scenes, but often overlook pedestrians and other non-vehicle dynamic actors, hindering a complete pipeline for dynamic urban scene reconstruction. To that end, we propose a comprehensive 3DGS framework for driving scenes, named OmniRe, that allows for accurate, full-length reconstruction of diverse dynamic objects in a driving log. OmniRe builds dynamic neural scene graphs based on Gaussian representations and constructs multiple local canonical spaces that model various dynamic actors, including vehicles, pedestrians, and cyclists, among many others. This capability is unmatched by existing methods. OmniRe allows us to holistically reconstruct different objects present in the scene, subsequently enabling the simulation of reconstructed scenarios with all actors participating in real-time (~60Hz). Extensive evaluations on the Waymo dataset show that our approach outperforms prior state-of-the-art methods quantitatively and qualitatively by a large margin. We believe our work fills a critical gap in driving reconstruction.</description><identifier>DOI: 10.48550/arxiv.2408.16760</identifier><language>eng</language><subject>Computer Science - Computer Vision and Pattern Recognition</subject><creationdate>2024-08</creationdate><rights>http://creativecommons.org/licenses/by/4.0</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/2408.16760$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2408.16760$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Chen, Ziyu</creatorcontrib><creatorcontrib>Yang, Jiawei</creatorcontrib><creatorcontrib>Huang, Jiahui</creatorcontrib><creatorcontrib>de Lutio, Riccardo</creatorcontrib><creatorcontrib>Esturo, Janick Martinez</creatorcontrib><creatorcontrib>Ivanovic, Boris</creatorcontrib><creatorcontrib>Litany, Or</creatorcontrib><creatorcontrib>Gojcic, Zan</creatorcontrib><creatorcontrib>Fidler, Sanja</creatorcontrib><creatorcontrib>Pavone, Marco</creatorcontrib><creatorcontrib>Song, Li</creatorcontrib><creatorcontrib>Wang, Yue</creatorcontrib><title>OmniRe: Omni Urban Scene Reconstruction</title><description>We introduce OmniRe, a holistic approach for efficiently reconstructing high-fidelity dynamic urban scenes from on-device logs. Recent methods for modeling driving sequences using neural radiance fields or Gaussian Splatting have demonstrated the potential of reconstructing challenging dynamic scenes, but often overlook pedestrians and other non-vehicle dynamic actors, hindering a complete pipeline for dynamic urban scene reconstruction. To that end, we propose a comprehensive 3DGS framework for driving scenes, named OmniRe, that allows for accurate, full-length reconstruction of diverse dynamic objects in a driving log. OmniRe builds dynamic neural scene graphs based on Gaussian representations and constructs multiple local canonical spaces that model various dynamic actors, including vehicles, pedestrians, and cyclists, among many others. This capability is unmatched by existing methods. OmniRe allows us to holistically reconstruct different objects present in the scene, subsequently enabling the simulation of reconstructed scenarios with all actors participating in real-time (~60Hz). Extensive evaluations on the Waymo dataset show that our approach outperforms prior state-of-the-art methods quantitatively and qualitatively by a large margin. We believe our work fills a critical gap in driving reconstruction.</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>eNpjYJA0NNAzsTA1NdBPLKrILNMzMjGw0DM0Mzcz4GRQ98_NywxKtVIA0QqhRUmJeQrByal5qQpBqcn5ecUlRaXJJZn5eTwMrGmJOcWpvFCam0HezTXE2UMXbGR8QVFmbmJRZTzI6Hiw0caEVQAANCYtIA</recordid><startdate>20240829</startdate><enddate>20240829</enddate><creator>Chen, Ziyu</creator><creator>Yang, Jiawei</creator><creator>Huang, Jiahui</creator><creator>de Lutio, Riccardo</creator><creator>Esturo, Janick Martinez</creator><creator>Ivanovic, Boris</creator><creator>Litany, Or</creator><creator>Gojcic, Zan</creator><creator>Fidler, Sanja</creator><creator>Pavone, Marco</creator><creator>Song, Li</creator><creator>Wang, Yue</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20240829</creationdate><title>OmniRe: Omni Urban Scene Reconstruction</title><author>Chen, Ziyu ; Yang, Jiawei ; Huang, Jiahui ; de Lutio, Riccardo ; Esturo, Janick Martinez ; Ivanovic, Boris ; Litany, Or ; Gojcic, Zan ; Fidler, Sanja ; Pavone, Marco ; Song, Li ; Wang, Yue</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-arxiv_primary_2408_167603</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>Chen, Ziyu</creatorcontrib><creatorcontrib>Yang, Jiawei</creatorcontrib><creatorcontrib>Huang, Jiahui</creatorcontrib><creatorcontrib>de Lutio, Riccardo</creatorcontrib><creatorcontrib>Esturo, Janick Martinez</creatorcontrib><creatorcontrib>Ivanovic, Boris</creatorcontrib><creatorcontrib>Litany, Or</creatorcontrib><creatorcontrib>Gojcic, Zan</creatorcontrib><creatorcontrib>Fidler, Sanja</creatorcontrib><creatorcontrib>Pavone, Marco</creatorcontrib><creatorcontrib>Song, Li</creatorcontrib><creatorcontrib>Wang, Yue</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Chen, Ziyu</au><au>Yang, Jiawei</au><au>Huang, Jiahui</au><au>de Lutio, Riccardo</au><au>Esturo, Janick Martinez</au><au>Ivanovic, Boris</au><au>Litany, Or</au><au>Gojcic, Zan</au><au>Fidler, Sanja</au><au>Pavone, Marco</au><au>Song, Li</au><au>Wang, Yue</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>OmniRe: Omni Urban Scene Reconstruction</atitle><date>2024-08-29</date><risdate>2024</risdate><abstract>We introduce OmniRe, a holistic approach for efficiently reconstructing high-fidelity dynamic urban scenes from on-device logs. Recent methods for modeling driving sequences using neural radiance fields or Gaussian Splatting have demonstrated the potential of reconstructing challenging dynamic scenes, but often overlook pedestrians and other non-vehicle dynamic actors, hindering a complete pipeline for dynamic urban scene reconstruction. To that end, we propose a comprehensive 3DGS framework for driving scenes, named OmniRe, that allows for accurate, full-length reconstruction of diverse dynamic objects in a driving log. OmniRe builds dynamic neural scene graphs based on Gaussian representations and constructs multiple local canonical spaces that model various dynamic actors, including vehicles, pedestrians, and cyclists, among many others. This capability is unmatched by existing methods. OmniRe allows us to holistically reconstruct different objects present in the scene, subsequently enabling the simulation of reconstructed scenarios with all actors participating in real-time (~60Hz). Extensive evaluations on the Waymo dataset show that our approach outperforms prior state-of-the-art methods quantitatively and qualitatively by a large margin. We believe our work fills a critical gap in driving reconstruction.</abstract><doi>10.48550/arxiv.2408.16760</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2408.16760
ispartof
issn
language eng
recordid cdi_arxiv_primary_2408_16760
source arXiv.org
subjects Computer Science - Computer Vision and Pattern Recognition
title OmniRe: Omni Urban Scene Reconstruction
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-15T00%3A53%3A25IST&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=OmniRe:%20Omni%20Urban%20Scene%20Reconstruction&rft.au=Chen,%20Ziyu&rft.date=2024-08-29&rft_id=info:doi/10.48550/arxiv.2408.16760&rft_dat=%3Carxiv_GOX%3E2408_16760%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