Online and Real-Time Tracking in a Surveillance Scenario

Online and Real-Time Tracking in a Surveillance Scenario

This paper presents an approach for tracking in a surveillance scenario. Typical aspects for this scenario are a 24/7 operation with a static camera mounted above the height of a human with many objects or people. The Multiple Object Tracking Benchmark 20 (MOT20) reflects this scenario best. We can...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Urbann, Oliver, Bredtmann, Oliver, Otten, Maximilian, Richter, Jan-Philip, Bauer, Thilo, Zibriczky, David
Format: Artikel
Sprache:eng
Schlagworte:
Computer Science - Computer Vision and Pattern Recognition
Computer Science - Learning
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 Urbann, Oliver
Bredtmann, Oliver
Otten, Maximilian
Richter, Jan-Philip
Bauer, Thilo
Zibriczky, David
description This paper presents an approach for tracking in a surveillance scenario. Typical aspects for this scenario are a 24/7 operation with a static camera mounted above the height of a human with many objects or people. The Multiple Object Tracking Benchmark 20 (MOT20) reflects this scenario best. We can show that our approach is real-time capable on this benchmark and outperforms all other real-time capable approaches in HOTA, MOTA, and IDF1. We achieve this by contributing a fast Siamese network reformulated for linear runtime (instead of quadratic) to generate fingerprints from detections. Thus, it is possible to associate the detections to Kalman filters based on multiple tracking specific ratings: Cosine similarity of fingerprints, Intersection over Union, and pixel distance ratio in the image.
doi_str_mv 10.48550/arxiv.2106.01153
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2106_01153</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2106_01153</sourcerecordid><originalsourceid>FETCH-LOGICAL-a673-4e4809ceccce9353013c50c45f5e2ec61f67d7ae798949dbafa4db9c7a5cc4633</originalsourceid><addsrcrecordid>eNotj0tqwzAUALXJorg9QFfVBexK1s9aFtM2hYCh9t68PD8HEUUJCjXJ7UuTrmY3zDD2LEWlG2PEK-RLWKpaClsJKY16YE2XYkjEIU38myCWQzgQHzLgPqQdD4kD73_yQiFGSEi8R0qQw_GRrWaIZ3r6Z8GGj_ehXZeb7vOrfduUYJ0qNelGeCREJK-MElKhEajNbKgmtHK2bnJAzjde-2kLM-hp69GBQdRWqYK93LW39PGUwwHydfxbGG8L6hcw4EC9</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Online and Real-Time Tracking in a Surveillance Scenario</title><source>arXiv.org</source><creator>Urbann, Oliver ; Bredtmann, Oliver ; Otten, Maximilian ; Richter, Jan-Philip ; Bauer, Thilo ; Zibriczky, David</creator><creatorcontrib>Urbann, Oliver ; Bredtmann, Oliver ; Otten, Maximilian ; Richter, Jan-Philip ; Bauer, Thilo ; Zibriczky, David</creatorcontrib><description>This paper presents an approach for tracking in a surveillance scenario. Typical aspects for this scenario are a 24/7 operation with a static camera mounted above the height of a human with many objects or people. The Multiple Object Tracking Benchmark 20 (MOT20) reflects this scenario best. We can show that our approach is real-time capable on this benchmark and outperforms all other real-time capable approaches in HOTA, MOTA, and IDF1. We achieve this by contributing a fast Siamese network reformulated for linear runtime (instead of quadratic) to generate fingerprints from detections. Thus, it is possible to associate the detections to Kalman filters based on multiple tracking specific ratings: Cosine similarity of fingerprints, Intersection over Union, and pixel distance ratio in the image.</description><identifier>DOI: 10.48550/arxiv.2106.01153</identifier><language>eng</language><subject>Computer Science - Computer Vision and Pattern Recognition ; Computer Science - Learning</subject><creationdate>2021-06</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,776,881</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2106.01153$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2106.01153$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Urbann, Oliver</creatorcontrib><creatorcontrib>Bredtmann, Oliver</creatorcontrib><creatorcontrib>Otten, Maximilian</creatorcontrib><creatorcontrib>Richter, Jan-Philip</creatorcontrib><creatorcontrib>Bauer, Thilo</creatorcontrib><creatorcontrib>Zibriczky, David</creatorcontrib><title>Online and Real-Time Tracking in a Surveillance Scenario</title><description>This paper presents an approach for tracking in a surveillance scenario. Typical aspects for this scenario are a 24/7 operation with a static camera mounted above the height of a human with many objects or people. The Multiple Object Tracking Benchmark 20 (MOT20) reflects this scenario best. We can show that our approach is real-time capable on this benchmark and outperforms all other real-time capable approaches in HOTA, MOTA, and IDF1. We achieve this by contributing a fast Siamese network reformulated for linear runtime (instead of quadratic) to generate fingerprints from detections. Thus, it is possible to associate the detections to Kalman filters based on multiple tracking specific ratings: Cosine similarity of fingerprints, Intersection over Union, and pixel distance ratio in the image.</description><subject>Computer Science - Computer Vision and Pattern Recognition</subject><subject>Computer Science - Learning</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2021</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotj0tqwzAUALXJorg9QFfVBexK1s9aFtM2hYCh9t68PD8HEUUJCjXJ7UuTrmY3zDD2LEWlG2PEK-RLWKpaClsJKY16YE2XYkjEIU38myCWQzgQHzLgPqQdD4kD73_yQiFGSEi8R0qQw_GRrWaIZ3r6Z8GGj_ehXZeb7vOrfduUYJ0qNelGeCREJK-MElKhEajNbKgmtHK2bnJAzjde-2kLM-hp69GBQdRWqYK93LW39PGUwwHydfxbGG8L6hcw4EC9</recordid><startdate>20210602</startdate><enddate>20210602</enddate><creator>Urbann, Oliver</creator><creator>Bredtmann, Oliver</creator><creator>Otten, Maximilian</creator><creator>Richter, Jan-Philip</creator><creator>Bauer, Thilo</creator><creator>Zibriczky, David</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20210602</creationdate><title>Online and Real-Time Tracking in a Surveillance Scenario</title><author>Urbann, Oliver ; Bredtmann, Oliver ; Otten, Maximilian ; Richter, Jan-Philip ; Bauer, Thilo ; Zibriczky, David</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a673-4e4809ceccce9353013c50c45f5e2ec61f67d7ae798949dbafa4db9c7a5cc4633</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2021</creationdate><topic>Computer Science - Computer Vision and Pattern Recognition</topic><topic>Computer Science - Learning</topic><toplevel>online_resources</toplevel><creatorcontrib>Urbann, Oliver</creatorcontrib><creatorcontrib>Bredtmann, Oliver</creatorcontrib><creatorcontrib>Otten, Maximilian</creatorcontrib><creatorcontrib>Richter, Jan-Philip</creatorcontrib><creatorcontrib>Bauer, Thilo</creatorcontrib><creatorcontrib>Zibriczky, David</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Urbann, Oliver</au><au>Bredtmann, Oliver</au><au>Otten, Maximilian</au><au>Richter, Jan-Philip</au><au>Bauer, Thilo</au><au>Zibriczky, David</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Online and Real-Time Tracking in a Surveillance Scenario</atitle><date>2021-06-02</date><risdate>2021</risdate><abstract>This paper presents an approach for tracking in a surveillance scenario. Typical aspects for this scenario are a 24/7 operation with a static camera mounted above the height of a human with many objects or people. The Multiple Object Tracking Benchmark 20 (MOT20) reflects this scenario best. We can show that our approach is real-time capable on this benchmark and outperforms all other real-time capable approaches in HOTA, MOTA, and IDF1. We achieve this by contributing a fast Siamese network reformulated for linear runtime (instead of quadratic) to generate fingerprints from detections. Thus, it is possible to associate the detections to Kalman filters based on multiple tracking specific ratings: Cosine similarity of fingerprints, Intersection over Union, and pixel distance ratio in the image.</abstract><doi>10.48550/arxiv.2106.01153</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2106.01153
ispartof
issn
language eng
recordid cdi_arxiv_primary_2106_01153
source arXiv.org
subjects Computer Science - Computer Vision and Pattern Recognition
Computer Science - Learning
title Online and Real-Time Tracking in a Surveillance Scenario
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-01T17%3A32%3A38IST&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=Online%20and%20Real-Time%20Tracking%20in%20a%20Surveillance%20Scenario&rft.au=Urbann,%20Oliver&rft.date=2021-06-02&rft_id=info:doi/10.48550/arxiv.2106.01153&rft_dat=%3Carxiv_GOX%3E2106_01153%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