360 in the Wild: Dataset for Depth Prediction and View Synthesis

The large abundance of perspective camera datasets facilitated the emergence of novel learning-based strategies for various tasks, such as camera localization, single image depth estimation, or view synthesis. However, panoramic or omnidirectional image datasets, including essential information, suc...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Park, Kibaek, Rameau, Francois, Park, Jaesik, Kweon, In So
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 Park, Kibaek
Rameau, Francois
Park, Jaesik
Kweon, In So
description The large abundance of perspective camera datasets facilitated the emergence of novel learning-based strategies for various tasks, such as camera localization, single image depth estimation, or view synthesis. However, panoramic or omnidirectional image datasets, including essential information, such as pose and depth, are mostly made with synthetic scenes. In this work, we introduce a large scale 360$^{\circ}$ videos dataset in the wild. This dataset has been carefully scraped from the Internet and has been captured from various locations worldwide. Hence, this dataset exhibits very diversified environments (e.g., indoor and outdoor) and contexts (e.g., with and without moving objects). Each of the 25K images constituting our dataset is provided with its respective camera's pose and depth map. We illustrate the relevance of our dataset for two main tasks, namely, single image depth estimation and view synthesis.
doi_str_mv 10.48550/arxiv.2406.18898
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2406_18898</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2406_18898</sourcerecordid><originalsourceid>FETCH-arxiv_primary_2406_188983</originalsourceid><addsrcrecordid>eNpjYJA0NNAzsTA1NdBPLKrILNMzMjEw0zO0sLC04GRwMDYzUMjMUyjJSFUIz8xJsVJwSSxJLE4tUUjLL1JwSS0oyVAIKEpNyUwuyczPU0jMS1EIy0wtVwiuzANqKc4s5mFgTUvMKU7lhdLcDPJuriHOHrpgu-ILijJzE4sq40F2xoPtNCasAgAsFTVk</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>360 in the Wild: Dataset for Depth Prediction and View Synthesis</title><source>arXiv.org</source><creator>Park, Kibaek ; Rameau, Francois ; Park, Jaesik ; Kweon, In So</creator><creatorcontrib>Park, Kibaek ; Rameau, Francois ; Park, Jaesik ; Kweon, In So</creatorcontrib><description>The large abundance of perspective camera datasets facilitated the emergence of novel learning-based strategies for various tasks, such as camera localization, single image depth estimation, or view synthesis. However, panoramic or omnidirectional image datasets, including essential information, such as pose and depth, are mostly made with synthetic scenes. In this work, we introduce a large scale 360$^{\circ}$ videos dataset in the wild. This dataset has been carefully scraped from the Internet and has been captured from various locations worldwide. Hence, this dataset exhibits very diversified environments (e.g., indoor and outdoor) and contexts (e.g., with and without moving objects). Each of the 25K images constituting our dataset is provided with its respective camera's pose and depth map. We illustrate the relevance of our dataset for two main tasks, namely, single image depth estimation and view synthesis.</description><identifier>DOI: 10.48550/arxiv.2406.18898</identifier><language>eng</language><subject>Computer Science - Artificial Intelligence ; Computer Science - Computer Vision and Pattern Recognition</subject><creationdate>2024-06</creationdate><rights>http://creativecommons.org/licenses/by-nc-nd/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/2406.18898$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2406.18898$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Park, Kibaek</creatorcontrib><creatorcontrib>Rameau, Francois</creatorcontrib><creatorcontrib>Park, Jaesik</creatorcontrib><creatorcontrib>Kweon, In So</creatorcontrib><title>360 in the Wild: Dataset for Depth Prediction and View Synthesis</title><description>The large abundance of perspective camera datasets facilitated the emergence of novel learning-based strategies for various tasks, such as camera localization, single image depth estimation, or view synthesis. However, panoramic or omnidirectional image datasets, including essential information, such as pose and depth, are mostly made with synthetic scenes. In this work, we introduce a large scale 360$^{\circ}$ videos dataset in the wild. This dataset has been carefully scraped from the Internet and has been captured from various locations worldwide. Hence, this dataset exhibits very diversified environments (e.g., indoor and outdoor) and contexts (e.g., with and without moving objects). Each of the 25K images constituting our dataset is provided with its respective camera's pose and depth map. We illustrate the relevance of our dataset for two main tasks, namely, single image depth estimation and view synthesis.</description><subject>Computer Science - Artificial Intelligence</subject><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>eNpjYJA0NNAzsTA1NdBPLKrILNMzMjEw0zO0sLC04GRwMDYzUMjMUyjJSFUIz8xJsVJwSSxJLE4tUUjLL1JwSS0oyVAIKEpNyUwuyczPU0jMS1EIy0wtVwiuzANqKc4s5mFgTUvMKU7lhdLcDPJuriHOHrpgu-ILijJzE4sq40F2xoPtNCasAgAsFTVk</recordid><startdate>20240627</startdate><enddate>20240627</enddate><creator>Park, Kibaek</creator><creator>Rameau, Francois</creator><creator>Park, Jaesik</creator><creator>Kweon, In So</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20240627</creationdate><title>360 in the Wild: Dataset for Depth Prediction and View Synthesis</title><author>Park, Kibaek ; Rameau, Francois ; Park, Jaesik ; Kweon, In So</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-arxiv_primary_2406_188983</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Computer Science - Artificial Intelligence</topic><topic>Computer Science - Computer Vision and Pattern Recognition</topic><toplevel>online_resources</toplevel><creatorcontrib>Park, Kibaek</creatorcontrib><creatorcontrib>Rameau, Francois</creatorcontrib><creatorcontrib>Park, Jaesik</creatorcontrib><creatorcontrib>Kweon, In So</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Park, Kibaek</au><au>Rameau, Francois</au><au>Park, Jaesik</au><au>Kweon, In So</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>360 in the Wild: Dataset for Depth Prediction and View Synthesis</atitle><date>2024-06-27</date><risdate>2024</risdate><abstract>The large abundance of perspective camera datasets facilitated the emergence of novel learning-based strategies for various tasks, such as camera localization, single image depth estimation, or view synthesis. However, panoramic or omnidirectional image datasets, including essential information, such as pose and depth, are mostly made with synthetic scenes. In this work, we introduce a large scale 360$^{\circ}$ videos dataset in the wild. This dataset has been carefully scraped from the Internet and has been captured from various locations worldwide. Hence, this dataset exhibits very diversified environments (e.g., indoor and outdoor) and contexts (e.g., with and without moving objects). Each of the 25K images constituting our dataset is provided with its respective camera's pose and depth map. We illustrate the relevance of our dataset for two main tasks, namely, single image depth estimation and view synthesis.</abstract><doi>10.48550/arxiv.2406.18898</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2406.18898
ispartof
issn
language eng
recordid cdi_arxiv_primary_2406_18898
source arXiv.org
subjects Computer Science - Artificial Intelligence
Computer Science - Computer Vision and Pattern Recognition
title 360 in the Wild: Dataset for Depth Prediction and View Synthesis
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-08T06%3A49%3A08IST&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=360%20in%20the%20Wild:%20Dataset%20for%20Depth%20Prediction%20and%20View%20Synthesis&rft.au=Park,%20Kibaek&rft.date=2024-06-27&rft_id=info:doi/10.48550/arxiv.2406.18898&rft_dat=%3Carxiv_GOX%3E2406_18898%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