3-D Mediated Detection and Tracking in Wide Area Aerial Surveillance

We address the problem of tracking many moving objects in wide area aerial surveillance. We propose that using 3-D information significantly improves performance over standard state of the art trackers, which rely on 2-D stabilization. We present and contrast two 3-D mediated approaches. The first a...

Full description

Saved in:
Bibliographic Details
Main Authors: Bor-Jeng Chen, Medioni, Gerard
Format: Conference Proceeding
Language:English
Subjects:
Cameras
Estimation
Geometry
Optical imaging
Solid modeling
Tracking
Videos
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites
container_end_page 403
container_issue
container_start_page 396
container_title
container_volume
creator Bor-Jeng Chen
Medioni, Gerard
description We address the problem of tracking many moving objects in wide area aerial surveillance. We propose that using 3-D information significantly improves performance over standard state of the art trackers, which rely on 2-D stabilization. We present and contrast two 3-D mediated approaches. The first approach assumes a dense 3-D model has previously been obtained. Given the camera poses, we can predict the image flow between frames and perform pixel-level classification for detection. The second approach instead computes the image flow, and relies on the epipolar geometry constraint to distinguish object motion from parallax. Our experiments on real imagery show a significant improvement in probability of detection (from 36% to 77%/67%), false alarm rate (from 75% to 3%/4%), and a speedup of an order of magnitude for the tracker itself. Using explicit 3-D improves the results, at a higher computational cost.
doi_str_mv 10.1109/WACV.2015.59
format conference_proceeding
fullrecord <record><control><sourceid>ieee_6IE</sourceid><recordid>TN_cdi_ieee_primary_7045913</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>7045913</ieee_id><sourcerecordid>7045913</sourcerecordid><originalsourceid>FETCH-LOGICAL-i208t-3a382a1370257e2ee75972ca89e5825e906e8052945abe1c207e4d7f856c75d73</originalsourceid><addsrcrecordid>eNotzLtOwzAUAFCDQCKUbmws_gGH68eN7TFKeElFDBQ6Vsa-RYYQkBOQ-HuQYDrbYexUQi0l-PNN2z3WCiTW6PfYsTTW-6ZxGvdZpRqjhNdOHrBKIoJA6-GILafpBQCkN_aXivVa9PyWUg4zJd7TTHHO7yMPY-LrEuJrHp95HvkmJ-JtocBbKjkM_P6zfFEehjBGOmGHuzBMtPx3wR4uL9bdtVjdXd107UpkBW4WOmingtQWFFpSRBa9VTE4T-gUkoeGHKDyBsMTyajAkkl257CJFpPVC3b292Yi2n6U_BbK99aCQS-1_gHa8Elw</addsrcrecordid><sourcetype>Publisher</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>3-D Mediated Detection and Tracking in Wide Area Aerial Surveillance</title><source>IEEE Electronic Library (IEL) Conference Proceedings</source><creator>Bor-Jeng Chen ; Medioni, Gerard</creator><creatorcontrib>Bor-Jeng Chen ; Medioni, Gerard</creatorcontrib><description>We address the problem of tracking many moving objects in wide area aerial surveillance. We propose that using 3-D information significantly improves performance over standard state of the art trackers, which rely on 2-D stabilization. We present and contrast two 3-D mediated approaches. The first approach assumes a dense 3-D model has previously been obtained. Given the camera poses, we can predict the image flow between frames and perform pixel-level classification for detection. The second approach instead computes the image flow, and relies on the epipolar geometry constraint to distinguish object motion from parallax. Our experiments on real imagery show a significant improvement in probability of detection (from 36% to 77%/67%), false alarm rate (from 75% to 3%/4%), and a speedup of an order of magnitude for the tracker itself. Using explicit 3-D improves the results, at a higher computational cost.</description><identifier>ISSN: 1550-5790</identifier><identifier>EISSN: 2642-9381</identifier><identifier>EISBN: 1479966835</identifier><identifier>EISBN: 9781479966837</identifier><identifier>DOI: 10.1109/WACV.2015.59</identifier><identifier>CODEN: IEEPAD</identifier><language>eng</language><publisher>IEEE</publisher><subject>Cameras ; Estimation ; Geometry ; Optical imaging ; Solid modeling ; Tracking ; Videos</subject><ispartof>2015 IEEE Winter Conference on Applications of Computer Vision, 2015, p.396-403</ispartof><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/7045913$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>309,310,776,780,785,786,2052,27902,54530,54895,54907</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/7045913$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Bor-Jeng Chen</creatorcontrib><creatorcontrib>Medioni, Gerard</creatorcontrib><title>3-D Mediated Detection and Tracking in Wide Area Aerial Surveillance</title><title>2015 IEEE Winter Conference on Applications of Computer Vision</title><addtitle>WACV</addtitle><description>We address the problem of tracking many moving objects in wide area aerial surveillance. We propose that using 3-D information significantly improves performance over standard state of the art trackers, which rely on 2-D stabilization. We present and contrast two 3-D mediated approaches. The first approach assumes a dense 3-D model has previously been obtained. Given the camera poses, we can predict the image flow between frames and perform pixel-level classification for detection. The second approach instead computes the image flow, and relies on the epipolar geometry constraint to distinguish object motion from parallax. Our experiments on real imagery show a significant improvement in probability of detection (from 36% to 77%/67%), false alarm rate (from 75% to 3%/4%), and a speedup of an order of magnitude for the tracker itself. Using explicit 3-D improves the results, at a higher computational cost.</description><subject>Cameras</subject><subject>Estimation</subject><subject>Geometry</subject><subject>Optical imaging</subject><subject>Solid modeling</subject><subject>Tracking</subject><subject>Videos</subject><issn>1550-5790</issn><issn>2642-9381</issn><isbn>1479966835</isbn><isbn>9781479966837</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2015</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><recordid>eNotzLtOwzAUAFCDQCKUbmws_gGH68eN7TFKeElFDBQ6Vsa-RYYQkBOQ-HuQYDrbYexUQi0l-PNN2z3WCiTW6PfYsTTW-6ZxGvdZpRqjhNdOHrBKIoJA6-GILafpBQCkN_aXivVa9PyWUg4zJd7TTHHO7yMPY-LrEuJrHp95HvkmJ-JtocBbKjkM_P6zfFEehjBGOmGHuzBMtPx3wR4uL9bdtVjdXd107UpkBW4WOmingtQWFFpSRBa9VTE4T-gUkoeGHKDyBsMTyajAkkl257CJFpPVC3b292Yi2n6U_BbK99aCQS-1_gHa8Elw</recordid><startdate>20150101</startdate><enddate>20150101</enddate><creator>Bor-Jeng Chen</creator><creator>Medioni, Gerard</creator><general>IEEE</general><scope>6IE</scope><scope>6IL</scope><scope>CBEJK</scope><scope>RIE</scope><scope>RIL</scope></search><sort><creationdate>20150101</creationdate><title>3-D Mediated Detection and Tracking in Wide Area Aerial Surveillance</title><author>Bor-Jeng Chen ; Medioni, Gerard</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-i208t-3a382a1370257e2ee75972ca89e5825e906e8052945abe1c207e4d7f856c75d73</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2015</creationdate><topic>Cameras</topic><topic>Estimation</topic><topic>Geometry</topic><topic>Optical imaging</topic><topic>Solid modeling</topic><topic>Tracking</topic><topic>Videos</topic><toplevel>online_resources</toplevel><creatorcontrib>Bor-Jeng Chen</creatorcontrib><creatorcontrib>Medioni, Gerard</creatorcontrib><collection>IEEE Electronic Library (IEL) Conference Proceedings</collection><collection>IEEE Proceedings Order Plan All Online (POP All Online) 1998-present by volume</collection><collection>IEEE Xplore All Conference Proceedings</collection><collection>IEL</collection><collection>IEEE Proceedings Order Plans (POP All) 1998-Present</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Bor-Jeng Chen</au><au>Medioni, Gerard</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>3-D Mediated Detection and Tracking in Wide Area Aerial Surveillance</atitle><btitle>2015 IEEE Winter Conference on Applications of Computer Vision</btitle><stitle>WACV</stitle><date>2015-01-01</date><risdate>2015</risdate><spage>396</spage><epage>403</epage><pages>396-403</pages><issn>1550-5790</issn><eissn>2642-9381</eissn><eisbn>1479966835</eisbn><eisbn>9781479966837</eisbn><coden>IEEPAD</coden><abstract>We address the problem of tracking many moving objects in wide area aerial surveillance. We propose that using 3-D information significantly improves performance over standard state of the art trackers, which rely on 2-D stabilization. We present and contrast two 3-D mediated approaches. The first approach assumes a dense 3-D model has previously been obtained. Given the camera poses, we can predict the image flow between frames and perform pixel-level classification for detection. The second approach instead computes the image flow, and relies on the epipolar geometry constraint to distinguish object motion from parallax. Our experiments on real imagery show a significant improvement in probability of detection (from 36% to 77%/67%), false alarm rate (from 75% to 3%/4%), and a speedup of an order of magnitude for the tracker itself. Using explicit 3-D improves the results, at a higher computational cost.</abstract><pub>IEEE</pub><doi>10.1109/WACV.2015.59</doi><tpages>8</tpages></addata></record>
fulltext fulltext_linktorsrc
identifier ISSN: 1550-5790
ispartof 2015 IEEE Winter Conference on Applications of Computer Vision, 2015, p.396-403
issn 1550-5790
2642-9381
language eng
recordid cdi_ieee_primary_7045913
source IEEE Electronic Library (IEL) Conference Proceedings
subjects Cameras
Estimation
Geometry
Optical imaging
Solid modeling
Tracking
Videos
title 3-D Mediated Detection and Tracking in Wide Area Aerial Surveillance
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-10T02%3A40%3A04IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-ieee_6IE&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=proceeding&rft.atitle=3-D%20Mediated%20Detection%20and%20Tracking%20in%20Wide%20Area%20Aerial%20Surveillance&rft.btitle=2015%20IEEE%20Winter%20Conference%20on%20Applications%20of%20Computer%20Vision&rft.au=Bor-Jeng%20Chen&rft.date=2015-01-01&rft.spage=396&rft.epage=403&rft.pages=396-403&rft.issn=1550-5790&rft.eissn=2642-9381&rft.coden=IEEPAD&rft_id=info:doi/10.1109/WACV.2015.59&rft.eisbn=1479966835&rft.eisbn_list=9781479966837&rft_dat=%3Cieee_6IE%3E7045913%3C/ieee_6IE%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-i208t-3a382a1370257e2ee75972ca89e5825e906e8052945abe1c207e4d7f856c75d73%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_id=info:pmid/&rft_ieee_id=7045913&rfr_iscdi=true