Loci-Segmented: Improving Scene Segmentation Learning

Loci-Segmented: Improving Scene Segmentation Learning

Current slot-oriented approaches for compositional scene segmentation from images and videos rely on provided background information or slot assignments. We present a segmented location and identity tracking system, Loci-Segmented (Loci-s), which does not require either of this information. It learn...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Traub, Manuel, Becker, Frederic, Sauter, Adrian, Otte, Sebastian, Butz, Martin V
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 Traub, Manuel
Becker, Frederic
Sauter, Adrian
Otte, Sebastian
Butz, Martin V
description Current slot-oriented approaches for compositional scene segmentation from images and videos rely on provided background information or slot assignments. We present a segmented location and identity tracking system, Loci-Segmented (Loci-s), which does not require either of this information. It learns to dynamically segment scenes into interpretable background and slot-based object encodings, separating rgb, mask, location, and depth information for each. The results reveal largely superior video decomposition performance in the MOVi datasets and in another established dataset collection targeting scene segmentation. The system's well-interpretable, compositional latent encodings may serve as a foundation model for downstream tasks.
doi_str_mv 10.48550/arxiv.2310.10410
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2310_10410</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2310_10410</sourcerecordid><originalsourceid>FETCH-LOGICAL-a670-6138514a4f71750bab70d0ef29e2d5c3e18333dd5320e56ada3f1ec11842b9633</originalsourceid><addsrcrecordid>eNotjkGKwkAQRXszi0E9wKwmF4hWdXUn0d0gOgoBF3EfKumKNEw60oro7XXU1Yf_4PGU-kKYmsJamHG8-stU0-NAMAifypZD69NKDr2Es7hFsu2Pcbj4cEiqVoIkb8RnP4SkFI7hwcbqo-O_k0zeO1L79Wq_3KTl7ne7_ClTznJIM6TComHT5ZhbaLjJwYF0ei7a2ZYECyJyzpIGsRk7pg6lRSyMbuYZ0Uh9v7TP7voYfc_xVv_3189-ugNaxD7r</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Loci-Segmented: Improving Scene Segmentation Learning</title><source>arXiv.org</source><creator>Traub, Manuel ; Becker, Frederic ; Sauter, Adrian ; Otte, Sebastian ; Butz, Martin V</creator><creatorcontrib>Traub, Manuel ; Becker, Frederic ; Sauter, Adrian ; Otte, Sebastian ; Butz, Martin V</creatorcontrib><description>Current slot-oriented approaches for compositional scene segmentation from images and videos rely on provided background information or slot assignments. We present a segmented location and identity tracking system, Loci-Segmented (Loci-s), which does not require either of this information. It learns to dynamically segment scenes into interpretable background and slot-based object encodings, separating rgb, mask, location, and depth information for each. The results reveal largely superior video decomposition performance in the MOVi datasets and in another established dataset collection targeting scene segmentation. The system's well-interpretable, compositional latent encodings may serve as a foundation model for downstream tasks.</description><identifier>DOI: 10.48550/arxiv.2310.10410</identifier><language>eng</language><subject>Computer Science - Computer Vision and Pattern Recognition</subject><creationdate>2023-10</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/2310.10410$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2310.10410$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Traub, Manuel</creatorcontrib><creatorcontrib>Becker, Frederic</creatorcontrib><creatorcontrib>Sauter, Adrian</creatorcontrib><creatorcontrib>Otte, Sebastian</creatorcontrib><creatorcontrib>Butz, Martin V</creatorcontrib><title>Loci-Segmented: Improving Scene Segmentation Learning</title><description>Current slot-oriented approaches for compositional scene segmentation from images and videos rely on provided background information or slot assignments. We present a segmented location and identity tracking system, Loci-Segmented (Loci-s), which does not require either of this information. It learns to dynamically segment scenes into interpretable background and slot-based object encodings, separating rgb, mask, location, and depth information for each. The results reveal largely superior video decomposition performance in the MOVi datasets and in another established dataset collection targeting scene segmentation. The system's well-interpretable, compositional latent encodings may serve as a foundation model for downstream tasks.</description><subject>Computer Science - Computer Vision and Pattern Recognition</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotjkGKwkAQRXszi0E9wKwmF4hWdXUn0d0gOgoBF3EfKumKNEw60oro7XXU1Yf_4PGU-kKYmsJamHG8-stU0-NAMAifypZD69NKDr2Es7hFsu2Pcbj4cEiqVoIkb8RnP4SkFI7hwcbqo-O_k0zeO1L79Wq_3KTl7ne7_ClTznJIM6TComHT5ZhbaLjJwYF0ei7a2ZYECyJyzpIGsRk7pg6lRSyMbuYZ0Uh9v7TP7voYfc_xVv_3189-ugNaxD7r</recordid><startdate>20231016</startdate><enddate>20231016</enddate><creator>Traub, Manuel</creator><creator>Becker, Frederic</creator><creator>Sauter, Adrian</creator><creator>Otte, Sebastian</creator><creator>Butz, Martin V</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20231016</creationdate><title>Loci-Segmented: Improving Scene Segmentation Learning</title><author>Traub, Manuel ; Becker, Frederic ; Sauter, Adrian ; Otte, Sebastian ; Butz, Martin V</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a670-6138514a4f71750bab70d0ef29e2d5c3e18333dd5320e56ada3f1ec11842b9633</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Computer Science - Computer Vision and Pattern Recognition</topic><toplevel>online_resources</toplevel><creatorcontrib>Traub, Manuel</creatorcontrib><creatorcontrib>Becker, Frederic</creatorcontrib><creatorcontrib>Sauter, Adrian</creatorcontrib><creatorcontrib>Otte, Sebastian</creatorcontrib><creatorcontrib>Butz, Martin V</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Traub, Manuel</au><au>Becker, Frederic</au><au>Sauter, Adrian</au><au>Otte, Sebastian</au><au>Butz, Martin V</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Loci-Segmented: Improving Scene Segmentation Learning</atitle><date>2023-10-16</date><risdate>2023</risdate><abstract>Current slot-oriented approaches for compositional scene segmentation from images and videos rely on provided background information or slot assignments. We present a segmented location and identity tracking system, Loci-Segmented (Loci-s), which does not require either of this information. It learns to dynamically segment scenes into interpretable background and slot-based object encodings, separating rgb, mask, location, and depth information for each. The results reveal largely superior video decomposition performance in the MOVi datasets and in another established dataset collection targeting scene segmentation. The system's well-interpretable, compositional latent encodings may serve as a foundation model for downstream tasks.</abstract><doi>10.48550/arxiv.2310.10410</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2310.10410
ispartof
issn
language eng
recordid cdi_arxiv_primary_2310_10410
source arXiv.org
subjects Computer Science - Computer Vision and Pattern Recognition
title Loci-Segmented: Improving Scene Segmentation Learning
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-27T05%3A24%3A00IST&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=Loci-Segmented:%20Improving%20Scene%20Segmentation%20Learning&rft.au=Traub,%20Manuel&rft.date=2023-10-16&rft_id=info:doi/10.48550/arxiv.2310.10410&rft_dat=%3Carxiv_GOX%3E2310_10410%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