Atrous Space Bender U-Net (ASBU-Net/LogiNet)

$ $With recent advances in CNNs, exceptional improvements have been made in semantic segmentation of high resolution images in terms of accuracy and latency. However, challenges still remain in detecting objects in crowded scenes, large scale variations, partial occlusion, and distortions, while sti...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Bansal, Anurag, Ostap, Oleg, Trueba, Miguel Maestre, Perry, Kristopher
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 Bansal, Anurag
Ostap, Oleg
Trueba, Miguel Maestre
Perry, Kristopher
description $ $With recent advances in CNNs, exceptional improvements have been made in semantic segmentation of high resolution images in terms of accuracy and latency. However, challenges still remain in detecting objects in crowded scenes, large scale variations, partial occlusion, and distortions, while still maintaining mobility and latency. We introduce a fast and efficient convolutional neural network, ASBU-Net, for semantic segmentation of high resolution images that addresses these problems and uses no novelty layers for ease of quantization and embedded hardware support. ASBU-Net is based on a new feature extraction module, atrous space bender layer (ASBL), which is efficient in terms of computation and memory. The ASB layers form a building block that is used to make ASBNet. Since this network does not use any special layers it can be easily implemented, quantized and deployed on FPGAs and other hardware with limited memory. We present experiments on resource and accuracy trade-offs and show strong performance compared to other popular models.
doi_str_mv 10.48550/arxiv.2212.08613
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2212_08613</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2212_08613</sourcerecordid><originalsourceid>FETCH-LOGICAL-a673-74ed620dedf1077067d1f47cfd63348efedc019ac63f627a72018874baf971103</originalsourceid><addsrcrecordid>eNotzj1vwjAUhWEvDAj4AUxkLBIJ99rG14wBQUGK6ACdIze-RpH4koGq_HvalOk909EjRB8h03YygbGLP_V3JiXKDKxB1Raj_BbP92uyvbiKkxmfPMfkM93wLXnLt7NmjYvzvv7tsCtawR2u3Hu1I3bLxW6-SouP9_U8L1JnSKWk2RsJnn1AIAJDHoOmKnijlLYc2FeAU1cZFYwkRxLQWtJfLkwJEVRHDP5vG255ifXRxUf5xy4btnoCwtM6EA</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Atrous Space Bender U-Net (ASBU-Net/LogiNet)</title><source>arXiv.org</source><creator>Bansal, Anurag ; Ostap, Oleg ; Trueba, Miguel Maestre ; Perry, Kristopher</creator><creatorcontrib>Bansal, Anurag ; Ostap, Oleg ; Trueba, Miguel Maestre ; Perry, Kristopher</creatorcontrib><description>$ $With recent advances in CNNs, exceptional improvements have been made in semantic segmentation of high resolution images in terms of accuracy and latency. However, challenges still remain in detecting objects in crowded scenes, large scale variations, partial occlusion, and distortions, while still maintaining mobility and latency. We introduce a fast and efficient convolutional neural network, ASBU-Net, for semantic segmentation of high resolution images that addresses these problems and uses no novelty layers for ease of quantization and embedded hardware support. ASBU-Net is based on a new feature extraction module, atrous space bender layer (ASBL), which is efficient in terms of computation and memory. The ASB layers form a building block that is used to make ASBNet. Since this network does not use any special layers it can be easily implemented, quantized and deployed on FPGAs and other hardware with limited memory. We present experiments on resource and accuracy trade-offs and show strong performance compared to other popular models.</description><identifier>DOI: 10.48550/arxiv.2212.08613</identifier><language>eng</language><subject>Computer Science - Computer Vision and Pattern Recognition</subject><creationdate>2022-12</creationdate><rights>http://creativecommons.org/licenses/by-sa/4.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,778,883</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2212.08613$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2212.08613$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Bansal, Anurag</creatorcontrib><creatorcontrib>Ostap, Oleg</creatorcontrib><creatorcontrib>Trueba, Miguel Maestre</creatorcontrib><creatorcontrib>Perry, Kristopher</creatorcontrib><title>Atrous Space Bender U-Net (ASBU-Net/LogiNet)</title><description>$ $With recent advances in CNNs, exceptional improvements have been made in semantic segmentation of high resolution images in terms of accuracy and latency. However, challenges still remain in detecting objects in crowded scenes, large scale variations, partial occlusion, and distortions, while still maintaining mobility and latency. We introduce a fast and efficient convolutional neural network, ASBU-Net, for semantic segmentation of high resolution images that addresses these problems and uses no novelty layers for ease of quantization and embedded hardware support. ASBU-Net is based on a new feature extraction module, atrous space bender layer (ASBL), which is efficient in terms of computation and memory. The ASB layers form a building block that is used to make ASBNet. Since this network does not use any special layers it can be easily implemented, quantized and deployed on FPGAs and other hardware with limited memory. We present experiments on resource and accuracy trade-offs and show strong performance compared to other popular models.</description><subject>Computer Science - Computer Vision and Pattern Recognition</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2022</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotzj1vwjAUhWEvDAj4AUxkLBIJ99rG14wBQUGK6ACdIze-RpH4koGq_HvalOk909EjRB8h03YygbGLP_V3JiXKDKxB1Raj_BbP92uyvbiKkxmfPMfkM93wLXnLt7NmjYvzvv7tsCtawR2u3Hu1I3bLxW6-SouP9_U8L1JnSKWk2RsJnn1AIAJDHoOmKnijlLYc2FeAU1cZFYwkRxLQWtJfLkwJEVRHDP5vG255ifXRxUf5xy4btnoCwtM6EA</recordid><startdate>20221216</startdate><enddate>20221216</enddate><creator>Bansal, Anurag</creator><creator>Ostap, Oleg</creator><creator>Trueba, Miguel Maestre</creator><creator>Perry, Kristopher</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20221216</creationdate><title>Atrous Space Bender U-Net (ASBU-Net/LogiNet)</title><author>Bansal, Anurag ; Ostap, Oleg ; Trueba, Miguel Maestre ; Perry, Kristopher</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a673-74ed620dedf1077067d1f47cfd63348efedc019ac63f627a72018874baf971103</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2022</creationdate><topic>Computer Science - Computer Vision and Pattern Recognition</topic><toplevel>online_resources</toplevel><creatorcontrib>Bansal, Anurag</creatorcontrib><creatorcontrib>Ostap, Oleg</creatorcontrib><creatorcontrib>Trueba, Miguel Maestre</creatorcontrib><creatorcontrib>Perry, Kristopher</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Bansal, Anurag</au><au>Ostap, Oleg</au><au>Trueba, Miguel Maestre</au><au>Perry, Kristopher</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Atrous Space Bender U-Net (ASBU-Net/LogiNet)</atitle><date>2022-12-16</date><risdate>2022</risdate><abstract>$ $With recent advances in CNNs, exceptional improvements have been made in semantic segmentation of high resolution images in terms of accuracy and latency. However, challenges still remain in detecting objects in crowded scenes, large scale variations, partial occlusion, and distortions, while still maintaining mobility and latency. We introduce a fast and efficient convolutional neural network, ASBU-Net, for semantic segmentation of high resolution images that addresses these problems and uses no novelty layers for ease of quantization and embedded hardware support. ASBU-Net is based on a new feature extraction module, atrous space bender layer (ASBL), which is efficient in terms of computation and memory. The ASB layers form a building block that is used to make ASBNet. Since this network does not use any special layers it can be easily implemented, quantized and deployed on FPGAs and other hardware with limited memory. We present experiments on resource and accuracy trade-offs and show strong performance compared to other popular models.</abstract><doi>10.48550/arxiv.2212.08613</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2212.08613
ispartof
issn
language eng
recordid cdi_arxiv_primary_2212_08613
source arXiv.org
subjects Computer Science - Computer Vision and Pattern Recognition
title Atrous Space Bender U-Net (ASBU-Net/LogiNet)
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-16T09%3A16%3A44IST&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=Atrous%20Space%20Bender%20U-Net%20(ASBU-Net/LogiNet)&rft.au=Bansal,%20Anurag&rft.date=2022-12-16&rft_id=info:doi/10.48550/arxiv.2212.08613&rft_dat=%3Carxiv_GOX%3E2212_08613%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