Loading…
Uncertainty Inspired RGB-D Saliency Detection
We propose the first stochastic framework to employ uncertainty for RGB-D saliency detection by learning from the data labeling process. Existing RGB-D saliency detection models treat this task as a point estimation problem by predicting a single saliency map following a deterministic learning pipel...
Saved in:
Published in: | IEEE transactions on pattern analysis and machine intelligence 2022-09, Vol.44 (9), p.5761-5779 |
---|---|
Main Authors: | , , , , , , |
Format: | Article |
Language: | English |
Subjects: | |
Citations: | Items that this one cites Items that cite this one |
Online Access: | Get full text |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
cited_by | cdi_FETCH-LOGICAL-c351t-5265778d3e9fae9c607d9efdcafdfe3a3c713e8e0a44451ea888e3b6e20b642c3 |
---|---|
cites | cdi_FETCH-LOGICAL-c351t-5265778d3e9fae9c607d9efdcafdfe3a3c713e8e0a44451ea888e3b6e20b642c3 |
container_end_page | 5779 |
container_issue | 9 |
container_start_page | 5761 |
container_title | IEEE transactions on pattern analysis and machine intelligence |
container_volume | 44 |
creator | Zhang, Jing Fan, Deng-Ping Dai, Yuchao Anwar, Saeed Saleh, Fatemeh Aliakbarian, Sadegh Barnes, Nick |
description | We propose the first stochastic framework to employ uncertainty for RGB-D saliency detection by learning from the data labeling process. Existing RGB-D saliency detection models treat this task as a point estimation problem by predicting a single saliency map following a deterministic learning pipeline. We argue that, however, the deterministic solution is relatively ill-posed. Inspired by the saliency data labeling process, we propose a generative architecture to achieve probabilistic RGB-D saliency detection which utilizes a latent variable to model the labeling variations. Our framework includes two main models: 1) a generator model, which maps the input image and latent variable to stochastic saliency prediction, and 2) an inference model, which gradually updates the latent variable by sampling it from the true or approximate posterior distribution. The generator model is an encoder-decoder saliency network. To infer the latent variable, we introduce two different solutions: i) a Conditional Variational Auto-encoder with an extra encoder to approximate the posterior distribution of the latent variable; and ii) an Alternating Back-Propagation technique, which directly samples the latent variable from the true posterior distribution. Qualitative and quantitative results on six challenging RGB-D benchmark datasets show our approach's superior performance in learning the distribution of saliency maps. The source code is publicly available via our project page: https://github.com/JingZhang617/UCNet . |
doi_str_mv | 10.1109/TPAMI.2021.3073564 |
format | article |
fullrecord | <record><control><sourceid>proquest_ieee_</sourceid><recordid>TN_cdi_ieee_primary_9405467</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>9405467</ieee_id><sourcerecordid>2698828134</sourcerecordid><originalsourceid>FETCH-LOGICAL-c351t-5265778d3e9fae9c607d9efdcafdfe3a3c713e8e0a44451ea888e3b6e20b642c3</originalsourceid><addsrcrecordid>eNpdkLFOwzAQhi0EoqXwAiChSCwsKbbPTuwRWiiVikDQzpbrXKRUaVLsZOjbk9LSgemG-_7T_R8h14wOGaP6Yf7x-DYdcsrZEGgKMhEnpM806Bgk6FPSpyzhsVJc9chFCCtKmZAUzkkPQMlEK94n8aJy6BtbVM02mlZhU3jMos_JUzyOvmxZYOW20RgbdE1RV5fkLLdlwKvDHJDFy_N89BrP3ifT0eMsdiBZE0ueyDRVGaDOLWqX0DTTmGfO5lmOYMGlDFAhtUIIydAqpRCWCXK6TAR3MCD3-7sbX3-3GBqzLoLDsrQV1m0wXHZNNNUs7dC7f-iqbn3VfWd4V7Erz0B0FN9TztcheMzNxhdr67eGUbOTaX5lmp1Mc5DZhW4Pp9vlGrNj5M9eB9zsgQIRj2stqBRJCj9aGHan</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2698828134</pqid></control><display><type>article</type><title>Uncertainty Inspired RGB-D Saliency Detection</title><source>IEEE Electronic Library (IEL) Journals</source><creator>Zhang, Jing ; Fan, Deng-Ping ; Dai, Yuchao ; Anwar, Saeed ; Saleh, Fatemeh ; Aliakbarian, Sadegh ; Barnes, Nick</creator><creatorcontrib>Zhang, Jing ; Fan, Deng-Ping ; Dai, Yuchao ; Anwar, Saeed ; Saleh, Fatemeh ; Aliakbarian, Sadegh ; Barnes, Nick</creatorcontrib><description>We propose the first stochastic framework to employ uncertainty for RGB-D saliency detection by learning from the data labeling process. Existing RGB-D saliency detection models treat this task as a point estimation problem by predicting a single saliency map following a deterministic learning pipeline. We argue that, however, the deterministic solution is relatively ill-posed. Inspired by the saliency data labeling process, we propose a generative architecture to achieve probabilistic RGB-D saliency detection which utilizes a latent variable to model the labeling variations. Our framework includes two main models: 1) a generator model, which maps the input image and latent variable to stochastic saliency prediction, and 2) an inference model, which gradually updates the latent variable by sampling it from the true or approximate posterior distribution. The generator model is an encoder-decoder saliency network. To infer the latent variable, we introduce two different solutions: i) a Conditional Variational Auto-encoder with an extra encoder to approximate the posterior distribution of the latent variable; and ii) an Alternating Back-Propagation technique, which directly samples the latent variable from the true posterior distribution. Qualitative and quantitative results on six challenging RGB-D benchmark datasets show our approach's superior performance in learning the distribution of saliency maps. The source code is publicly available via our project page: https://github.com/JingZhang617/UCNet .</description><identifier>ISSN: 0162-8828</identifier><identifier>EISSN: 1939-3539</identifier><identifier>EISSN: 2160-9292</identifier><identifier>DOI: 10.1109/TPAMI.2021.3073564</identifier><identifier>PMID: 33856982</identifier><identifier>CODEN: ITPIDJ</identifier><language>eng</language><publisher>United States: IEEE</publisher><subject>alternating back-propagation ; Back propagation ; Coders ; conditional variational autoencoders ; Data models ; Encoders-Decoders ; Labeling ; Labelling ; Learning ; Pipelines ; Predictive models ; RGB-D saliency detection ; Salience ; Saliency detection ; Source code ; Training ; Uncertainty</subject><ispartof>IEEE transactions on pattern analysis and machine intelligence, 2022-09, Vol.44 (9), p.5761-5779</ispartof><rights>Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2022</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c351t-5265778d3e9fae9c607d9efdcafdfe3a3c713e8e0a44451ea888e3b6e20b642c3</citedby><cites>FETCH-LOGICAL-c351t-5265778d3e9fae9c607d9efdcafdfe3a3c713e8e0a44451ea888e3b6e20b642c3</cites><orcidid>0000-0002-9343-9535 ; 0000-0002-5245-7518 ; 0000-0002-4432-7406 ; 0000-0002-0692-8411</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/9405467$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>315,786,790,27957,27958,55147</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/33856982$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Zhang, Jing</creatorcontrib><creatorcontrib>Fan, Deng-Ping</creatorcontrib><creatorcontrib>Dai, Yuchao</creatorcontrib><creatorcontrib>Anwar, Saeed</creatorcontrib><creatorcontrib>Saleh, Fatemeh</creatorcontrib><creatorcontrib>Aliakbarian, Sadegh</creatorcontrib><creatorcontrib>Barnes, Nick</creatorcontrib><title>Uncertainty Inspired RGB-D Saliency Detection</title><title>IEEE transactions on pattern analysis and machine intelligence</title><addtitle>TPAMI</addtitle><addtitle>IEEE Trans Pattern Anal Mach Intell</addtitle><description>We propose the first stochastic framework to employ uncertainty for RGB-D saliency detection by learning from the data labeling process. Existing RGB-D saliency detection models treat this task as a point estimation problem by predicting a single saliency map following a deterministic learning pipeline. We argue that, however, the deterministic solution is relatively ill-posed. Inspired by the saliency data labeling process, we propose a generative architecture to achieve probabilistic RGB-D saliency detection which utilizes a latent variable to model the labeling variations. Our framework includes two main models: 1) a generator model, which maps the input image and latent variable to stochastic saliency prediction, and 2) an inference model, which gradually updates the latent variable by sampling it from the true or approximate posterior distribution. The generator model is an encoder-decoder saliency network. To infer the latent variable, we introduce two different solutions: i) a Conditional Variational Auto-encoder with an extra encoder to approximate the posterior distribution of the latent variable; and ii) an Alternating Back-Propagation technique, which directly samples the latent variable from the true posterior distribution. Qualitative and quantitative results on six challenging RGB-D benchmark datasets show our approach's superior performance in learning the distribution of saliency maps. The source code is publicly available via our project page: https://github.com/JingZhang617/UCNet .</description><subject>alternating back-propagation</subject><subject>Back propagation</subject><subject>Coders</subject><subject>conditional variational autoencoders</subject><subject>Data models</subject><subject>Encoders-Decoders</subject><subject>Labeling</subject><subject>Labelling</subject><subject>Learning</subject><subject>Pipelines</subject><subject>Predictive models</subject><subject>RGB-D saliency detection</subject><subject>Salience</subject><subject>Saliency detection</subject><subject>Source code</subject><subject>Training</subject><subject>Uncertainty</subject><issn>0162-8828</issn><issn>1939-3539</issn><issn>2160-9292</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2022</creationdate><recordtype>article</recordtype><recordid>eNpdkLFOwzAQhi0EoqXwAiChSCwsKbbPTuwRWiiVikDQzpbrXKRUaVLsZOjbk9LSgemG-_7T_R8h14wOGaP6Yf7x-DYdcsrZEGgKMhEnpM806Bgk6FPSpyzhsVJc9chFCCtKmZAUzkkPQMlEK94n8aJy6BtbVM02mlZhU3jMos_JUzyOvmxZYOW20RgbdE1RV5fkLLdlwKvDHJDFy_N89BrP3ifT0eMsdiBZE0ueyDRVGaDOLWqX0DTTmGfO5lmOYMGlDFAhtUIIydAqpRCWCXK6TAR3MCD3-7sbX3-3GBqzLoLDsrQV1m0wXHZNNNUs7dC7f-iqbn3VfWd4V7Erz0B0FN9TztcheMzNxhdr67eGUbOTaX5lmp1Mc5DZhW4Pp9vlGrNj5M9eB9zsgQIRj2stqBRJCj9aGHan</recordid><startdate>20220901</startdate><enddate>20220901</enddate><creator>Zhang, Jing</creator><creator>Fan, Deng-Ping</creator><creator>Dai, Yuchao</creator><creator>Anwar, Saeed</creator><creator>Saleh, Fatemeh</creator><creator>Aliakbarian, Sadegh</creator><creator>Barnes, Nick</creator><general>IEEE</general><general>The Institute of Electrical and Electronics Engineers, Inc. (IEEE)</general><scope>97E</scope><scope>RIA</scope><scope>RIE</scope><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7SC</scope><scope>7SP</scope><scope>8FD</scope><scope>JQ2</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope><scope>7X8</scope><orcidid>https://orcid.org/0000-0002-9343-9535</orcidid><orcidid>https://orcid.org/0000-0002-5245-7518</orcidid><orcidid>https://orcid.org/0000-0002-4432-7406</orcidid><orcidid>https://orcid.org/0000-0002-0692-8411</orcidid></search><sort><creationdate>20220901</creationdate><title>Uncertainty Inspired RGB-D Saliency Detection</title><author>Zhang, Jing ; Fan, Deng-Ping ; Dai, Yuchao ; Anwar, Saeed ; Saleh, Fatemeh ; Aliakbarian, Sadegh ; Barnes, Nick</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c351t-5265778d3e9fae9c607d9efdcafdfe3a3c713e8e0a44451ea888e3b6e20b642c3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2022</creationdate><topic>alternating back-propagation</topic><topic>Back propagation</topic><topic>Coders</topic><topic>conditional variational autoencoders</topic><topic>Data models</topic><topic>Encoders-Decoders</topic><topic>Labeling</topic><topic>Labelling</topic><topic>Learning</topic><topic>Pipelines</topic><topic>Predictive models</topic><topic>RGB-D saliency detection</topic><topic>Salience</topic><topic>Saliency detection</topic><topic>Source code</topic><topic>Training</topic><topic>Uncertainty</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Zhang, Jing</creatorcontrib><creatorcontrib>Fan, Deng-Ping</creatorcontrib><creatorcontrib>Dai, Yuchao</creatorcontrib><creatorcontrib>Anwar, Saeed</creatorcontrib><creatorcontrib>Saleh, Fatemeh</creatorcontrib><creatorcontrib>Aliakbarian, Sadegh</creatorcontrib><creatorcontrib>Barnes, Nick</creatorcontrib><collection>IEEE All-Society Periodicals Package (ASPP) 2005-present</collection><collection>IEEE All-Society Periodicals Package (ASPP) 1998–Present</collection><collection>IEEE Xplore</collection><collection>PubMed</collection><collection>CrossRef</collection><collection>Computer and Information Systems Abstracts</collection><collection>Electronics & Communications Abstracts</collection><collection>Technology Research Database</collection><collection>ProQuest Computer Science Collection</collection><collection>Advanced Technologies Database with Aerospace</collection><collection>Computer and Information Systems Abstracts Academic</collection><collection>Computer and Information Systems Abstracts Professional</collection><collection>MEDLINE - Academic</collection><jtitle>IEEE transactions on pattern analysis and machine intelligence</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Zhang, Jing</au><au>Fan, Deng-Ping</au><au>Dai, Yuchao</au><au>Anwar, Saeed</au><au>Saleh, Fatemeh</au><au>Aliakbarian, Sadegh</au><au>Barnes, Nick</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Uncertainty Inspired RGB-D Saliency Detection</atitle><jtitle>IEEE transactions on pattern analysis and machine intelligence</jtitle><stitle>TPAMI</stitle><addtitle>IEEE Trans Pattern Anal Mach Intell</addtitle><date>2022-09-01</date><risdate>2022</risdate><volume>44</volume><issue>9</issue><spage>5761</spage><epage>5779</epage><pages>5761-5779</pages><issn>0162-8828</issn><eissn>1939-3539</eissn><eissn>2160-9292</eissn><coden>ITPIDJ</coden><notes>ObjectType-Article-1</notes><notes>SourceType-Scholarly Journals-1</notes><notes>ObjectType-Feature-2</notes><notes>content type line 23</notes><abstract>We propose the first stochastic framework to employ uncertainty for RGB-D saliency detection by learning from the data labeling process. Existing RGB-D saliency detection models treat this task as a point estimation problem by predicting a single saliency map following a deterministic learning pipeline. We argue that, however, the deterministic solution is relatively ill-posed. Inspired by the saliency data labeling process, we propose a generative architecture to achieve probabilistic RGB-D saliency detection which utilizes a latent variable to model the labeling variations. Our framework includes two main models: 1) a generator model, which maps the input image and latent variable to stochastic saliency prediction, and 2) an inference model, which gradually updates the latent variable by sampling it from the true or approximate posterior distribution. The generator model is an encoder-decoder saliency network. To infer the latent variable, we introduce two different solutions: i) a Conditional Variational Auto-encoder with an extra encoder to approximate the posterior distribution of the latent variable; and ii) an Alternating Back-Propagation technique, which directly samples the latent variable from the true posterior distribution. Qualitative and quantitative results on six challenging RGB-D benchmark datasets show our approach's superior performance in learning the distribution of saliency maps. The source code is publicly available via our project page: https://github.com/JingZhang617/UCNet .</abstract><cop>United States</cop><pub>IEEE</pub><pmid>33856982</pmid><doi>10.1109/TPAMI.2021.3073564</doi><tpages>19</tpages><orcidid>https://orcid.org/0000-0002-9343-9535</orcidid><orcidid>https://orcid.org/0000-0002-5245-7518</orcidid><orcidid>https://orcid.org/0000-0002-4432-7406</orcidid><orcidid>https://orcid.org/0000-0002-0692-8411</orcidid></addata></record> |
fulltext | fulltext |
identifier | ISSN: 0162-8828 |
ispartof | IEEE transactions on pattern analysis and machine intelligence, 2022-09, Vol.44 (9), p.5761-5779 |
issn | 0162-8828 1939-3539 2160-9292 |
language | eng |
recordid | cdi_ieee_primary_9405467 |
source | IEEE Electronic Library (IEL) Journals |
subjects | alternating back-propagation Back propagation Coders conditional variational autoencoders Data models Encoders-Decoders Labeling Labelling Learning Pipelines Predictive models RGB-D saliency detection Salience Saliency detection Source code Training Uncertainty |
title | Uncertainty Inspired RGB-D Saliency Detection |
url | http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-09-22T04%3A57%3A48IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_ieee_&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Uncertainty%20Inspired%20RGB-D%20Saliency%20Detection&rft.jtitle=IEEE%20transactions%20on%20pattern%20analysis%20and%20machine%20intelligence&rft.au=Zhang,%20Jing&rft.date=2022-09-01&rft.volume=44&rft.issue=9&rft.spage=5761&rft.epage=5779&rft.pages=5761-5779&rft.issn=0162-8828&rft.eissn=1939-3539&rft.coden=ITPIDJ&rft_id=info:doi/10.1109/TPAMI.2021.3073564&rft_dat=%3Cproquest_ieee_%3E2698828134%3C/proquest_ieee_%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c351t-5265778d3e9fae9c607d9efdcafdfe3a3c713e8e0a44451ea888e3b6e20b642c3%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2698828134&rft_id=info:pmid/33856982&rft_ieee_id=9405467&rfr_iscdi=true |