PROCESSING IMAGES USING SELF-ATTENTION BASED NEURAL NETWORKS

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for processing images using self-attention based neural networks. One of the methods includes obtaining one or more images comprising a plurality of pixels; determining, for each image of the one or more...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Dosovitskiy, Alexey, Heigold, Georg, Beyer, Lucas Klaus, Weissenborn, Dirk, Zhai, Xiaohua, Gelly, Sylvain, Kolesnikov, Alexander, Dehghani, Mostafa, Houlsby, Neil Matthew Tinmouth, Minderer, Matthias Johannes Lorenz, Unterthiner, Thomas, Uszkoreit, Jakob D
Format: Patent
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 Dosovitskiy, Alexey
Heigold, Georg
Beyer, Lucas Klaus
Weissenborn, Dirk
Zhai, Xiaohua
Gelly, Sylvain
Kolesnikov, Alexander
Dehghani, Mostafa
Houlsby, Neil Matthew Tinmouth
Minderer, Matthias Johannes Lorenz
Unterthiner, Thomas
Uszkoreit, Jakob D
description Methods, systems, and apparatus, including computer programs encoded on computer storage media, for processing images using self-attention based neural networks. One of the methods includes obtaining one or more images comprising a plurality of pixels; determining, for each image of the one or more images, a plurality of image patches of the image, wherein each image patch comprises a different subset of the pixels of the image; processing, for each image of the one or more images, the corresponding plurality of image patches to generate an input sequence comprising a respective input element at each of a plurality of input positions, wherein a plurality of the input elements correspond to respective different image patches; and processing the input sequences using a neural network to generate a network output that characterizes the one or more images, wherein the neural network comprises one or more self-attention neural network layers.
format Patent
fullrecord <record><control><sourceid>epo_EVB</sourceid><recordid>TN_cdi_epo_espacenet_US2024062426A1</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>US2024062426A1</sourcerecordid><originalsourceid>FETCH-epo_espacenet_US2024062426A13</originalsourceid><addsrcrecordid>eNrjZLAJCPJ3dg0O9vRzV_D0dXR3DVYIBXOCXX3cdB1DQlz9Qjz9_RScHINdXRT8XEODHH2AVEi4f5B3MA8Da1piTnEqL5TmZlB2cw1x9tBNLciPTy0uSExOzUstiQ8NNjIwMjEwMzIxMnM0NCZOFQDJMSq4</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>patent</recordtype></control><display><type>patent</type><title>PROCESSING IMAGES USING SELF-ATTENTION BASED NEURAL NETWORKS</title><source>esp@cenet</source><creator>Dosovitskiy, Alexey ; Heigold, Georg ; Beyer, Lucas Klaus ; Weissenborn, Dirk ; Zhai, Xiaohua ; Gelly, Sylvain ; Kolesnikov, Alexander ; Dehghani, Mostafa ; Houlsby, Neil Matthew Tinmouth ; Minderer, Matthias Johannes Lorenz ; Unterthiner, Thomas ; Uszkoreit, Jakob D</creator><creatorcontrib>Dosovitskiy, Alexey ; Heigold, Georg ; Beyer, Lucas Klaus ; Weissenborn, Dirk ; Zhai, Xiaohua ; Gelly, Sylvain ; Kolesnikov, Alexander ; Dehghani, Mostafa ; Houlsby, Neil Matthew Tinmouth ; Minderer, Matthias Johannes Lorenz ; Unterthiner, Thomas ; Uszkoreit, Jakob D</creatorcontrib><description>Methods, systems, and apparatus, including computer programs encoded on computer storage media, for processing images using self-attention based neural networks. One of the methods includes obtaining one or more images comprising a plurality of pixels; determining, for each image of the one or more images, a plurality of image patches of the image, wherein each image patch comprises a different subset of the pixels of the image; processing, for each image of the one or more images, the corresponding plurality of image patches to generate an input sequence comprising a respective input element at each of a plurality of input positions, wherein a plurality of the input elements correspond to respective different image patches; and processing the input sequences using a neural network to generate a network output that characterizes the one or more images, wherein the neural network comprises one or more self-attention neural network layers.</description><language>eng</language><subject>CALCULATING ; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS ; COMPUTING ; COUNTING ; ELECTRIC DIGITAL DATA PROCESSING ; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL ; PHYSICS</subject><creationdate>2024</creationdate><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&amp;date=20240222&amp;DB=EPODOC&amp;CC=US&amp;NR=2024062426A1$$EHTML$$P50$$Gepo$$Hfree_for_read</linktohtml><link.rule.ids>230,308,780,885,25564,76547</link.rule.ids><linktorsrc>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&amp;date=20240222&amp;DB=EPODOC&amp;CC=US&amp;NR=2024062426A1$$EView_record_in_European_Patent_Office$$FView_record_in_$$GEuropean_Patent_Office$$Hfree_for_read</linktorsrc></links><search><creatorcontrib>Dosovitskiy, Alexey</creatorcontrib><creatorcontrib>Heigold, Georg</creatorcontrib><creatorcontrib>Beyer, Lucas Klaus</creatorcontrib><creatorcontrib>Weissenborn, Dirk</creatorcontrib><creatorcontrib>Zhai, Xiaohua</creatorcontrib><creatorcontrib>Gelly, Sylvain</creatorcontrib><creatorcontrib>Kolesnikov, Alexander</creatorcontrib><creatorcontrib>Dehghani, Mostafa</creatorcontrib><creatorcontrib>Houlsby, Neil Matthew Tinmouth</creatorcontrib><creatorcontrib>Minderer, Matthias Johannes Lorenz</creatorcontrib><creatorcontrib>Unterthiner, Thomas</creatorcontrib><creatorcontrib>Uszkoreit, Jakob D</creatorcontrib><title>PROCESSING IMAGES USING SELF-ATTENTION BASED NEURAL NETWORKS</title><description>Methods, systems, and apparatus, including computer programs encoded on computer storage media, for processing images using self-attention based neural networks. One of the methods includes obtaining one or more images comprising a plurality of pixels; determining, for each image of the one or more images, a plurality of image patches of the image, wherein each image patch comprises a different subset of the pixels of the image; processing, for each image of the one or more images, the corresponding plurality of image patches to generate an input sequence comprising a respective input element at each of a plurality of input positions, wherein a plurality of the input elements correspond to respective different image patches; and processing the input sequences using a neural network to generate a network output that characterizes the one or more images, wherein the neural network comprises one or more self-attention neural network layers.</description><subject>CALCULATING</subject><subject>COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS</subject><subject>COMPUTING</subject><subject>COUNTING</subject><subject>ELECTRIC DIGITAL DATA PROCESSING</subject><subject>IMAGE DATA PROCESSING OR GENERATION, IN GENERAL</subject><subject>PHYSICS</subject><fulltext>true</fulltext><rsrctype>patent</rsrctype><creationdate>2024</creationdate><recordtype>patent</recordtype><sourceid>EVB</sourceid><recordid>eNrjZLAJCPJ3dg0O9vRzV_D0dXR3DVYIBXOCXX3cdB1DQlz9Qjz9_RScHINdXRT8XEODHH2AVEi4f5B3MA8Da1piTnEqL5TmZlB2cw1x9tBNLciPTy0uSExOzUstiQ8NNjIwMjEwMzIxMnM0NCZOFQDJMSq4</recordid><startdate>20240222</startdate><enddate>20240222</enddate><creator>Dosovitskiy, Alexey</creator><creator>Heigold, Georg</creator><creator>Beyer, Lucas Klaus</creator><creator>Weissenborn, Dirk</creator><creator>Zhai, Xiaohua</creator><creator>Gelly, Sylvain</creator><creator>Kolesnikov, Alexander</creator><creator>Dehghani, Mostafa</creator><creator>Houlsby, Neil Matthew Tinmouth</creator><creator>Minderer, Matthias Johannes Lorenz</creator><creator>Unterthiner, Thomas</creator><creator>Uszkoreit, Jakob D</creator><scope>EVB</scope></search><sort><creationdate>20240222</creationdate><title>PROCESSING IMAGES USING SELF-ATTENTION BASED NEURAL NETWORKS</title><author>Dosovitskiy, Alexey ; Heigold, Georg ; Beyer, Lucas Klaus ; Weissenborn, Dirk ; Zhai, Xiaohua ; Gelly, Sylvain ; Kolesnikov, Alexander ; Dehghani, Mostafa ; Houlsby, Neil Matthew Tinmouth ; Minderer, Matthias Johannes Lorenz ; Unterthiner, Thomas ; Uszkoreit, Jakob D</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-epo_espacenet_US2024062426A13</frbrgroupid><rsrctype>patents</rsrctype><prefilter>patents</prefilter><language>eng</language><creationdate>2024</creationdate><topic>CALCULATING</topic><topic>COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS</topic><topic>COMPUTING</topic><topic>COUNTING</topic><topic>ELECTRIC DIGITAL DATA PROCESSING</topic><topic>IMAGE DATA PROCESSING OR GENERATION, IN GENERAL</topic><topic>PHYSICS</topic><toplevel>online_resources</toplevel><creatorcontrib>Dosovitskiy, Alexey</creatorcontrib><creatorcontrib>Heigold, Georg</creatorcontrib><creatorcontrib>Beyer, Lucas Klaus</creatorcontrib><creatorcontrib>Weissenborn, Dirk</creatorcontrib><creatorcontrib>Zhai, Xiaohua</creatorcontrib><creatorcontrib>Gelly, Sylvain</creatorcontrib><creatorcontrib>Kolesnikov, Alexander</creatorcontrib><creatorcontrib>Dehghani, Mostafa</creatorcontrib><creatorcontrib>Houlsby, Neil Matthew Tinmouth</creatorcontrib><creatorcontrib>Minderer, Matthias Johannes Lorenz</creatorcontrib><creatorcontrib>Unterthiner, Thomas</creatorcontrib><creatorcontrib>Uszkoreit, Jakob D</creatorcontrib><collection>esp@cenet</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Dosovitskiy, Alexey</au><au>Heigold, Georg</au><au>Beyer, Lucas Klaus</au><au>Weissenborn, Dirk</au><au>Zhai, Xiaohua</au><au>Gelly, Sylvain</au><au>Kolesnikov, Alexander</au><au>Dehghani, Mostafa</au><au>Houlsby, Neil Matthew Tinmouth</au><au>Minderer, Matthias Johannes Lorenz</au><au>Unterthiner, Thomas</au><au>Uszkoreit, Jakob D</au><format>patent</format><genre>patent</genre><ristype>GEN</ristype><title>PROCESSING IMAGES USING SELF-ATTENTION BASED NEURAL NETWORKS</title><date>2024-02-22</date><risdate>2024</risdate><abstract>Methods, systems, and apparatus, including computer programs encoded on computer storage media, for processing images using self-attention based neural networks. One of the methods includes obtaining one or more images comprising a plurality of pixels; determining, for each image of the one or more images, a plurality of image patches of the image, wherein each image patch comprises a different subset of the pixels of the image; processing, for each image of the one or more images, the corresponding plurality of image patches to generate an input sequence comprising a respective input element at each of a plurality of input positions, wherein a plurality of the input elements correspond to respective different image patches; and processing the input sequences using a neural network to generate a network output that characterizes the one or more images, wherein the neural network comprises one or more self-attention neural network layers.</abstract><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier
ispartof
issn
language eng
recordid cdi_epo_espacenet_US2024062426A1
source esp@cenet
subjects CALCULATING
COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
COMPUTING
COUNTING
ELECTRIC DIGITAL DATA PROCESSING
IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
PHYSICS
title PROCESSING IMAGES USING SELF-ATTENTION BASED NEURAL NETWORKS
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-23T09%3A38%3A49IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-epo_EVB&rft_val_fmt=info:ofi/fmt:kev:mtx:patent&rft.genre=patent&rft.au=Dosovitskiy,%20Alexey&rft.date=2024-02-22&rft_id=info:doi/&rft_dat=%3Cepo_EVB%3EUS2024062426A1%3C/epo_EVB%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