TSDM: Tracking by SiamRPN++ with a Depth-refiner and a Mask-generator
In a generic object tracking, depth (D) information provides informative cues for foreground-background separation and target bounding box regression. However, so far, few trackers have used depth information to play the important role aforementioned due to the lack of a suitable model. In this pape...
Gespeichert in:
Hauptverfasser: | , , , |
---|---|
Format: | Artikel |
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 | Zhao, Pengyao Liu, Quanli Wang, Wei Guo, Qiang |
description | In a generic object tracking, depth (D) information provides informative cues
for foreground-background separation and target bounding box regression.
However, so far, few trackers have used depth information to play the important
role aforementioned due to the lack of a suitable model. In this paper, a RGB-D
tracker named TSDM is proposed, which is composed of a Mask-generator (M-g),
SiamRPN++ and a Depth-refiner (D-r). The M-g generates the background masks,
and updates them as the target 3D position changes. The D-r optimizes the
target bounding box estimated by SiamRPN++, based on the spatial depth
distribution difference between the target and the surrounding background.
Extensive evaluation on the Princeton Tracking Benchmark and the Visual Object
Tracking challenge shows that our tracker outperforms the state-of-the-art by a
large margin while achieving 23 FPS. In addition, a light-weight variant can
run at 31 FPS and thus it is practical for real world applications. Code and
models of TSDM are available at https://github.com/lql-team/TSDM. |
doi_str_mv | 10.48550/arxiv.2005.04063 |
format | Article |
fullrecord | <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2005_04063</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2005_04063</sourcerecordid><originalsourceid>FETCH-LOGICAL-a673-21074102ff7e6fa122ed8210ac9220df1c76aada8bd1e9b5fdd3f4f57661b3763</originalsourceid><addsrcrecordid>eNotz01PAjEUheFuWBjwB7iye9Lxtp22AzsC-JGAGpn95M60FxpkJGWi8u8d0dVJnsVJXsZuJGR5YQzcYfqOn5kCMBnkYPUVW5abxXrKy4TNPrZbXp_5JuLh7fV5POZfsdtx5Itw7HYiBYptSBxb39saT3uxDT1g95FGbED4fgrX_ztk5f2ynD-K1cvD03y2EmidFkqCyyUoIhcsoVQq-KJHbCZKgSfZOIvosai9DJPakPeacjLOWllrZ_WQ3f7dXjqqY4oHTOfqt6e69OgfJllDsQ</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>TSDM: Tracking by SiamRPN++ with a Depth-refiner and a Mask-generator</title><source>arXiv.org</source><creator>Zhao, Pengyao ; Liu, Quanli ; Wang, Wei ; Guo, Qiang</creator><creatorcontrib>Zhao, Pengyao ; Liu, Quanli ; Wang, Wei ; Guo, Qiang</creatorcontrib><description>In a generic object tracking, depth (D) information provides informative cues
for foreground-background separation and target bounding box regression.
However, so far, few trackers have used depth information to play the important
role aforementioned due to the lack of a suitable model. In this paper, a RGB-D
tracker named TSDM is proposed, which is composed of a Mask-generator (M-g),
SiamRPN++ and a Depth-refiner (D-r). The M-g generates the background masks,
and updates them as the target 3D position changes. The D-r optimizes the
target bounding box estimated by SiamRPN++, based on the spatial depth
distribution difference between the target and the surrounding background.
Extensive evaluation on the Princeton Tracking Benchmark and the Visual Object
Tracking challenge shows that our tracker outperforms the state-of-the-art by a
large margin while achieving 23 FPS. In addition, a light-weight variant can
run at 31 FPS and thus it is practical for real world applications. Code and
models of TSDM are available at https://github.com/lql-team/TSDM.</description><identifier>DOI: 10.48550/arxiv.2005.04063</identifier><language>eng</language><subject>Computer Science - Computer Vision and Pattern Recognition</subject><creationdate>2020-05</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>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,780,885</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2005.04063$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2005.04063$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Zhao, Pengyao</creatorcontrib><creatorcontrib>Liu, Quanli</creatorcontrib><creatorcontrib>Wang, Wei</creatorcontrib><creatorcontrib>Guo, Qiang</creatorcontrib><title>TSDM: Tracking by SiamRPN++ with a Depth-refiner and a Mask-generator</title><description>In a generic object tracking, depth (D) information provides informative cues
for foreground-background separation and target bounding box regression.
However, so far, few trackers have used depth information to play the important
role aforementioned due to the lack of a suitable model. In this paper, a RGB-D
tracker named TSDM is proposed, which is composed of a Mask-generator (M-g),
SiamRPN++ and a Depth-refiner (D-r). The M-g generates the background masks,
and updates them as the target 3D position changes. The D-r optimizes the
target bounding box estimated by SiamRPN++, based on the spatial depth
distribution difference between the target and the surrounding background.
Extensive evaluation on the Princeton Tracking Benchmark and the Visual Object
Tracking challenge shows that our tracker outperforms the state-of-the-art by a
large margin while achieving 23 FPS. In addition, a light-weight variant can
run at 31 FPS and thus it is practical for real world applications. Code and
models of TSDM are available at https://github.com/lql-team/TSDM.</description><subject>Computer Science - Computer Vision and Pattern Recognition</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2020</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotz01PAjEUheFuWBjwB7iye9Lxtp22AzsC-JGAGpn95M60FxpkJGWi8u8d0dVJnsVJXsZuJGR5YQzcYfqOn5kCMBnkYPUVW5abxXrKy4TNPrZbXp_5JuLh7fV5POZfsdtx5Itw7HYiBYptSBxb39saT3uxDT1g95FGbED4fgrX_ztk5f2ynD-K1cvD03y2EmidFkqCyyUoIhcsoVQq-KJHbCZKgSfZOIvosai9DJPakPeacjLOWllrZ_WQ3f7dXjqqY4oHTOfqt6e69OgfJllDsQ</recordid><startdate>20200508</startdate><enddate>20200508</enddate><creator>Zhao, Pengyao</creator><creator>Liu, Quanli</creator><creator>Wang, Wei</creator><creator>Guo, Qiang</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20200508</creationdate><title>TSDM: Tracking by SiamRPN++ with a Depth-refiner and a Mask-generator</title><author>Zhao, Pengyao ; Liu, Quanli ; Wang, Wei ; Guo, Qiang</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a673-21074102ff7e6fa122ed8210ac9220df1c76aada8bd1e9b5fdd3f4f57661b3763</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2020</creationdate><topic>Computer Science - Computer Vision and Pattern Recognition</topic><toplevel>online_resources</toplevel><creatorcontrib>Zhao, Pengyao</creatorcontrib><creatorcontrib>Liu, Quanli</creatorcontrib><creatorcontrib>Wang, Wei</creatorcontrib><creatorcontrib>Guo, Qiang</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Zhao, Pengyao</au><au>Liu, Quanli</au><au>Wang, Wei</au><au>Guo, Qiang</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>TSDM: Tracking by SiamRPN++ with a Depth-refiner and a Mask-generator</atitle><date>2020-05-08</date><risdate>2020</risdate><abstract>In a generic object tracking, depth (D) information provides informative cues
for foreground-background separation and target bounding box regression.
However, so far, few trackers have used depth information to play the important
role aforementioned due to the lack of a suitable model. In this paper, a RGB-D
tracker named TSDM is proposed, which is composed of a Mask-generator (M-g),
SiamRPN++ and a Depth-refiner (D-r). The M-g generates the background masks,
and updates them as the target 3D position changes. The D-r optimizes the
target bounding box estimated by SiamRPN++, based on the spatial depth
distribution difference between the target and the surrounding background.
Extensive evaluation on the Princeton Tracking Benchmark and the Visual Object
Tracking challenge shows that our tracker outperforms the state-of-the-art by a
large margin while achieving 23 FPS. In addition, a light-weight variant can
run at 31 FPS and thus it is practical for real world applications. Code and
models of TSDM are available at https://github.com/lql-team/TSDM.</abstract><doi>10.48550/arxiv.2005.04063</doi><oa>free_for_read</oa></addata></record> |
fulltext | fulltext_linktorsrc |
identifier | DOI: 10.48550/arxiv.2005.04063 |
ispartof | |
issn | |
language | eng |
recordid | cdi_arxiv_primary_2005_04063 |
source | arXiv.org |
subjects | Computer Science - Computer Vision and Pattern Recognition |
title | TSDM: Tracking by SiamRPN++ with a Depth-refiner and a Mask-generator |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-25T10%3A27%3A39IST&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=TSDM:%20Tracking%20by%20SiamRPN++%20with%20a%20Depth-refiner%20and%20a%20Mask-generator&rft.au=Zhao,%20Pengyao&rft.date=2020-05-08&rft_id=info:doi/10.48550/arxiv.2005.04063&rft_dat=%3Carxiv_GOX%3E2005_04063%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 |