Efficient Uncertainty Estimation for Semantic Segmentation in Videos

Uncertainty estimation in deep learning becomes more important recently. A deep learning model can't be applied in real applications if we don't know whether the model is certain about the decision or not. Some literature proposes the Bayesian neural network which can estimate the uncertai...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:arXiv.org 2018-07
Hauptverfasser: Po-Yu, Huang, Wan-Ting, Hsu, Chun-Yueh Chiu, Ting-Fan, Wu, Sun, Min
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title arXiv.org
container_volume
creator Po-Yu, Huang
Wan-Ting, Hsu
Chun-Yueh Chiu
Ting-Fan, Wu
Sun, Min
description Uncertainty estimation in deep learning becomes more important recently. A deep learning model can't be applied in real applications if we don't know whether the model is certain about the decision or not. Some literature proposes the Bayesian neural network which can estimate the uncertainty by Monte Carlo Dropout (MC dropout). However, MC dropout needs to forward the model \(N\) times which results in \(N\) times slower. For real-time applications such as a self-driving car system, which needs to obtain the prediction and the uncertainty as fast as possible, so that MC dropout becomes impractical. In this work, we propose the region-based temporal aggregation (RTA) method which leverages the temporal information in videos to simulate the sampling procedure. Our RTA method with Tiramisu backbone is 10x faster than the MC dropout with Tiramisu backbone (\(N=5\)). Furthermore, the uncertainty estimation obtained by our RTA method is comparable to MC dropout's uncertainty estimation on pixel-level and frame-level metrics.
format Article
fullrecord <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2092770106</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2092770106</sourcerecordid><originalsourceid>FETCH-proquest_journals_20927701063</originalsourceid><addsrcrecordid>eNqNissKwjAQAIMgWLT_EPBcSDd96Fkr3n1cS6iJbLEbTbYH_96CfoCnGZiZiQS0zrNNAbAQaYy9UgqqGspSJ2LfOIcdWmJ5oc4GNkj8lk1kHAyjJ-l8kCc7GGLsJrkP0_stSPKKN-vjSsydeUSb_rgU60Nz3h2zZ_Cv0UZuez8GmlILagt1rXJV6f-uD2wAOtc</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2092770106</pqid></control><display><type>article</type><title>Efficient Uncertainty Estimation for Semantic Segmentation in Videos</title><source>Free E- Journals</source><creator>Po-Yu, Huang ; Wan-Ting, Hsu ; Chun-Yueh Chiu ; Ting-Fan, Wu ; Sun, Min</creator><creatorcontrib>Po-Yu, Huang ; Wan-Ting, Hsu ; Chun-Yueh Chiu ; Ting-Fan, Wu ; Sun, Min</creatorcontrib><description>Uncertainty estimation in deep learning becomes more important recently. A deep learning model can't be applied in real applications if we don't know whether the model is certain about the decision or not. Some literature proposes the Bayesian neural network which can estimate the uncertainty by Monte Carlo Dropout (MC dropout). However, MC dropout needs to forward the model \(N\) times which results in \(N\) times slower. For real-time applications such as a self-driving car system, which needs to obtain the prediction and the uncertainty as fast as possible, so that MC dropout becomes impractical. In this work, we propose the region-based temporal aggregation (RTA) method which leverages the temporal information in videos to simulate the sampling procedure. Our RTA method with Tiramisu backbone is 10x faster than the MC dropout with Tiramisu backbone (\(N=5\)). Furthermore, the uncertainty estimation obtained by our RTA method is comparable to MC dropout's uncertainty estimation on pixel-level and frame-level metrics.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Autonomous cars ; Backbone ; Bayesian analysis ; Computer simulation ; Machine learning ; Neural networks ; Semantic segmentation ; Uncertainty</subject><ispartof>arXiv.org, 2018-07</ispartof><rights>2018. This work is published under http://arxiv.org/licenses/nonexclusive-distrib/1.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</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>781,785</link.rule.ids></links><search><creatorcontrib>Po-Yu, Huang</creatorcontrib><creatorcontrib>Wan-Ting, Hsu</creatorcontrib><creatorcontrib>Chun-Yueh Chiu</creatorcontrib><creatorcontrib>Ting-Fan, Wu</creatorcontrib><creatorcontrib>Sun, Min</creatorcontrib><title>Efficient Uncertainty Estimation for Semantic Segmentation in Videos</title><title>arXiv.org</title><description>Uncertainty estimation in deep learning becomes more important recently. A deep learning model can't be applied in real applications if we don't know whether the model is certain about the decision or not. Some literature proposes the Bayesian neural network which can estimate the uncertainty by Monte Carlo Dropout (MC dropout). However, MC dropout needs to forward the model \(N\) times which results in \(N\) times slower. For real-time applications such as a self-driving car system, which needs to obtain the prediction and the uncertainty as fast as possible, so that MC dropout becomes impractical. In this work, we propose the region-based temporal aggregation (RTA) method which leverages the temporal information in videos to simulate the sampling procedure. Our RTA method with Tiramisu backbone is 10x faster than the MC dropout with Tiramisu backbone (\(N=5\)). Furthermore, the uncertainty estimation obtained by our RTA method is comparable to MC dropout's uncertainty estimation on pixel-level and frame-level metrics.</description><subject>Autonomous cars</subject><subject>Backbone</subject><subject>Bayesian analysis</subject><subject>Computer simulation</subject><subject>Machine learning</subject><subject>Neural networks</subject><subject>Semantic segmentation</subject><subject>Uncertainty</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2018</creationdate><recordtype>article</recordtype><sourceid>ABUWG</sourceid><sourceid>AFKRA</sourceid><sourceid>AZQEC</sourceid><sourceid>BENPR</sourceid><sourceid>CCPQU</sourceid><sourceid>DWQXO</sourceid><recordid>eNqNissKwjAQAIMgWLT_EPBcSDd96Fkr3n1cS6iJbLEbTbYH_96CfoCnGZiZiQS0zrNNAbAQaYy9UgqqGspSJ2LfOIcdWmJ5oc4GNkj8lk1kHAyjJ-l8kCc7GGLsJrkP0_stSPKKN-vjSsydeUSb_rgU60Nz3h2zZ_Cv0UZuez8GmlILagt1rXJV6f-uD2wAOtc</recordid><startdate>20180729</startdate><enddate>20180729</enddate><creator>Po-Yu, Huang</creator><creator>Wan-Ting, Hsu</creator><creator>Chun-Yueh Chiu</creator><creator>Ting-Fan, Wu</creator><creator>Sun, Min</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope></search><sort><creationdate>20180729</creationdate><title>Efficient Uncertainty Estimation for Semantic Segmentation in Videos</title><author>Po-Yu, Huang ; Wan-Ting, Hsu ; Chun-Yueh Chiu ; Ting-Fan, Wu ; Sun, Min</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_20927701063</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2018</creationdate><topic>Autonomous cars</topic><topic>Backbone</topic><topic>Bayesian analysis</topic><topic>Computer simulation</topic><topic>Machine learning</topic><topic>Neural networks</topic><topic>Semantic segmentation</topic><topic>Uncertainty</topic><toplevel>online_resources</toplevel><creatorcontrib>Po-Yu, Huang</creatorcontrib><creatorcontrib>Wan-Ting, Hsu</creatorcontrib><creatorcontrib>Chun-Yueh Chiu</creatorcontrib><creatorcontrib>Ting-Fan, Wu</creatorcontrib><creatorcontrib>Sun, Min</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science &amp; Engineering Collection</collection><collection>ProQuest Central (Alumni Edition)</collection><collection>ProQuest Central UK/Ireland</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>Publicly Available Content Database</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering Collection</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Po-Yu, Huang</au><au>Wan-Ting, Hsu</au><au>Chun-Yueh Chiu</au><au>Ting-Fan, Wu</au><au>Sun, Min</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>Efficient Uncertainty Estimation for Semantic Segmentation in Videos</atitle><jtitle>arXiv.org</jtitle><date>2018-07-29</date><risdate>2018</risdate><eissn>2331-8422</eissn><abstract>Uncertainty estimation in deep learning becomes more important recently. A deep learning model can't be applied in real applications if we don't know whether the model is certain about the decision or not. Some literature proposes the Bayesian neural network which can estimate the uncertainty by Monte Carlo Dropout (MC dropout). However, MC dropout needs to forward the model \(N\) times which results in \(N\) times slower. For real-time applications such as a self-driving car system, which needs to obtain the prediction and the uncertainty as fast as possible, so that MC dropout becomes impractical. In this work, we propose the region-based temporal aggregation (RTA) method which leverages the temporal information in videos to simulate the sampling procedure. Our RTA method with Tiramisu backbone is 10x faster than the MC dropout with Tiramisu backbone (\(N=5\)). Furthermore, the uncertainty estimation obtained by our RTA method is comparable to MC dropout's uncertainty estimation on pixel-level and frame-level metrics.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier EISSN: 2331-8422
ispartof arXiv.org, 2018-07
issn 2331-8422
language eng
recordid cdi_proquest_journals_2092770106
source Free E- Journals
subjects Autonomous cars
Backbone
Bayesian analysis
Computer simulation
Machine learning
Neural networks
Semantic segmentation
Uncertainty
title Efficient Uncertainty Estimation for Semantic Segmentation in Videos
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-18T11%3A12%3A58IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=document&rft.atitle=Efficient%20Uncertainty%20Estimation%20for%20Semantic%20Segmentation%20in%20Videos&rft.jtitle=arXiv.org&rft.au=Po-Yu,%20Huang&rft.date=2018-07-29&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2092770106%3C/proquest%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2092770106&rft_id=info:pmid/&rfr_iscdi=true