3D Scene Parsing via Class-Wise Adaptation

We propose the method that uses only computer graphics datasets to parse the real world 3D scenes. 3D scene parsing based on semantic segmentation is required to implement the categorical interaction in the virtual world. Convolutional Neural Networks (CNNs) have recently shown state-of-theart perfo...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Ono, Daichi, Yabe, Hiroyuki, Horikawa, Tsutomu
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 Ono, Daichi
Yabe, Hiroyuki
Horikawa, Tsutomu
description We propose the method that uses only computer graphics datasets to parse the real world 3D scenes. 3D scene parsing based on semantic segmentation is required to implement the categorical interaction in the virtual world. Convolutional Neural Networks (CNNs) have recently shown state-of-theart performance on computer vision tasks including semantic segmentation. However, collecting and annotating a huge amount of data are needed to train CNNs. Especially in the case of semantic segmentation, annotating pixel by pixel takes a significant amount of time and often makes mistakes. In contrast, computer graphics can generate a lot of accurate annotated data and easily scale up by changing camera positions, textures and lights. Despite these advantages, models trained on computer graphics datasets cannot perform well on real data, which is known as the domain shift. To address this issue, we first present that depth modal and synthetic noise are effective to reduce the domain shift. Then, we develop the class-wise adaptation which obtains domain invariant features of CNNs. To reduce the domain shift, we create computer graphics rooms with a lot of props, and provide photo-realistic rendered images.We also demonstrate the application which is combined semantic segmentation with Simultaneous Localization and Mapping (SLAM). Our application performs accurate 3D scene parsing in real-time on an actual room.
doi_str_mv 10.48550/arxiv.1812.03622
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_1812_03622</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>1812_03622</sourcerecordid><originalsourceid>FETCH-LOGICAL-a672-8dcc47783a77f16536e7cdf8a5593ab55bb44d6653a961759e4a63e82b4946643</originalsourceid><addsrcrecordid>eNotzjsLwjAYheEsDqL-ACczC63N7Us6lnqFgoKCY_naphKoVRoR_fdepzO8cHgIGbMolEapaIbdw91DZhgPIwGc98lUzOm-tK2lO-y8a0_07pCmDXofHJ23NKnwesObu7RD0qux8Xb03wE5LBeHdB1k29UmTbIAQfPAVGUptTYCta4ZKAFWl1VtUKlYYKFUUUhZwTtgDEyr2EoEYQ0vZCwBpBiQye_2i82vnTtj98w_6PyLFi-d3jnl</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>3D Scene Parsing via Class-Wise Adaptation</title><source>arXiv.org</source><creator>Ono, Daichi ; Yabe, Hiroyuki ; Horikawa, Tsutomu</creator><creatorcontrib>Ono, Daichi ; Yabe, Hiroyuki ; Horikawa, Tsutomu</creatorcontrib><description>We propose the method that uses only computer graphics datasets to parse the real world 3D scenes. 3D scene parsing based on semantic segmentation is required to implement the categorical interaction in the virtual world. Convolutional Neural Networks (CNNs) have recently shown state-of-theart performance on computer vision tasks including semantic segmentation. However, collecting and annotating a huge amount of data are needed to train CNNs. Especially in the case of semantic segmentation, annotating pixel by pixel takes a significant amount of time and often makes mistakes. In contrast, computer graphics can generate a lot of accurate annotated data and easily scale up by changing camera positions, textures and lights. Despite these advantages, models trained on computer graphics datasets cannot perform well on real data, which is known as the domain shift. To address this issue, we first present that depth modal and synthetic noise are effective to reduce the domain shift. Then, we develop the class-wise adaptation which obtains domain invariant features of CNNs. To reduce the domain shift, we create computer graphics rooms with a lot of props, and provide photo-realistic rendered images.We also demonstrate the application which is combined semantic segmentation with Simultaneous Localization and Mapping (SLAM). Our application performs accurate 3D scene parsing in real-time on an actual room.</description><identifier>DOI: 10.48550/arxiv.1812.03622</identifier><language>eng</language><subject>Computer Science - Computer Vision and Pattern Recognition</subject><creationdate>2018-12</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/1812.03622$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.1812.03622$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Ono, Daichi</creatorcontrib><creatorcontrib>Yabe, Hiroyuki</creatorcontrib><creatorcontrib>Horikawa, Tsutomu</creatorcontrib><title>3D Scene Parsing via Class-Wise Adaptation</title><description>We propose the method that uses only computer graphics datasets to parse the real world 3D scenes. 3D scene parsing based on semantic segmentation is required to implement the categorical interaction in the virtual world. Convolutional Neural Networks (CNNs) have recently shown state-of-theart performance on computer vision tasks including semantic segmentation. However, collecting and annotating a huge amount of data are needed to train CNNs. Especially in the case of semantic segmentation, annotating pixel by pixel takes a significant amount of time and often makes mistakes. In contrast, computer graphics can generate a lot of accurate annotated data and easily scale up by changing camera positions, textures and lights. Despite these advantages, models trained on computer graphics datasets cannot perform well on real data, which is known as the domain shift. To address this issue, we first present that depth modal and synthetic noise are effective to reduce the domain shift. Then, we develop the class-wise adaptation which obtains domain invariant features of CNNs. To reduce the domain shift, we create computer graphics rooms with a lot of props, and provide photo-realistic rendered images.We also demonstrate the application which is combined semantic segmentation with Simultaneous Localization and Mapping (SLAM). Our application performs accurate 3D scene parsing in real-time on an actual room.</description><subject>Computer Science - Computer Vision and Pattern Recognition</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2018</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotzjsLwjAYheEsDqL-ACczC63N7Us6lnqFgoKCY_naphKoVRoR_fdepzO8cHgIGbMolEapaIbdw91DZhgPIwGc98lUzOm-tK2lO-y8a0_07pCmDXofHJ23NKnwesObu7RD0qux8Xb03wE5LBeHdB1k29UmTbIAQfPAVGUptTYCta4ZKAFWl1VtUKlYYKFUUUhZwTtgDEyr2EoEYQ0vZCwBpBiQye_2i82vnTtj98w_6PyLFi-d3jnl</recordid><startdate>20181209</startdate><enddate>20181209</enddate><creator>Ono, Daichi</creator><creator>Yabe, Hiroyuki</creator><creator>Horikawa, Tsutomu</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20181209</creationdate><title>3D Scene Parsing via Class-Wise Adaptation</title><author>Ono, Daichi ; Yabe, Hiroyuki ; Horikawa, Tsutomu</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a672-8dcc47783a77f16536e7cdf8a5593ab55bb44d6653a961759e4a63e82b4946643</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2018</creationdate><topic>Computer Science - Computer Vision and Pattern Recognition</topic><toplevel>online_resources</toplevel><creatorcontrib>Ono, Daichi</creatorcontrib><creatorcontrib>Yabe, Hiroyuki</creatorcontrib><creatorcontrib>Horikawa, Tsutomu</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Ono, Daichi</au><au>Yabe, Hiroyuki</au><au>Horikawa, Tsutomu</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>3D Scene Parsing via Class-Wise Adaptation</atitle><date>2018-12-09</date><risdate>2018</risdate><abstract>We propose the method that uses only computer graphics datasets to parse the real world 3D scenes. 3D scene parsing based on semantic segmentation is required to implement the categorical interaction in the virtual world. Convolutional Neural Networks (CNNs) have recently shown state-of-theart performance on computer vision tasks including semantic segmentation. However, collecting and annotating a huge amount of data are needed to train CNNs. Especially in the case of semantic segmentation, annotating pixel by pixel takes a significant amount of time and often makes mistakes. In contrast, computer graphics can generate a lot of accurate annotated data and easily scale up by changing camera positions, textures and lights. Despite these advantages, models trained on computer graphics datasets cannot perform well on real data, which is known as the domain shift. To address this issue, we first present that depth modal and synthetic noise are effective to reduce the domain shift. Then, we develop the class-wise adaptation which obtains domain invariant features of CNNs. To reduce the domain shift, we create computer graphics rooms with a lot of props, and provide photo-realistic rendered images.We also demonstrate the application which is combined semantic segmentation with Simultaneous Localization and Mapping (SLAM). Our application performs accurate 3D scene parsing in real-time on an actual room.</abstract><doi>10.48550/arxiv.1812.03622</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.1812.03622
ispartof
issn
language eng
recordid cdi_arxiv_primary_1812_03622
source arXiv.org
subjects Computer Science - Computer Vision and Pattern Recognition
title 3D Scene Parsing via Class-Wise Adaptation
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-09T04%3A27%3A12IST&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=3D%20Scene%20Parsing%20via%20Class-Wise%20Adaptation&rft.au=Ono,%20Daichi&rft.date=2018-12-09&rft_id=info:doi/10.48550/arxiv.1812.03622&rft_dat=%3Carxiv_GOX%3E1812_03622%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