Journal article 1297 views 521 downloads
Transformation of an Uncertain Video Search Pipeline to a Sketch-Based Visual Analytics Loop
Philip A Legg,
David H. S Chung,
Matthew L Parry,
Rhodri Bown,
Mark Jones
,
Iwan W Griffiths,
Min Chen
IEEE Transactions on Visualization and Computer Graphics, Volume: 19, Issue: 12, Pages: 2109 - 2118
Swansea University Author:
Mark Jones
-
PDF | Version of Record
Download (12.88MB)
DOI (Published version): 10.1109/TVCG.2013.207
Abstract
Traditional sketch-based image or video search systems rely on machine learning concepts as their core technology. However, in many applications, machine learning alone is impractical since videos may not be semantically annotated sufficiently, there may be a lack of suitable training data, and the...
Published in: | IEEE Transactions on Visualization and Computer Graphics |
---|---|
Published: |
2013
|
URI: | https://cronfa.swan.ac.uk/Record/cronfa16829 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
first_indexed |
2014-01-30T17:01:10Z |
---|---|
last_indexed |
2019-06-21T13:24:44Z |
id |
cronfa16829 |
recordtype |
SURis |
fullrecord |
<?xml version="1.0"?><rfc1807><datestamp>2019-06-21T13:02:05.1385895</datestamp><bib-version>v2</bib-version><id>16829</id><entry>2014-01-09</entry><title>Transformation of an Uncertain Video Search Pipeline to a Sketch-Based Visual Analytics Loop</title><swanseaauthors><author><sid>2e1030b6e14fc9debd5d5ae7cc335562</sid><ORCID>0000-0001-8991-1190</ORCID><firstname>Mark</firstname><surname>Jones</surname><name>Mark Jones</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2014-01-09</date><deptcode>SCS</deptcode><abstract>Traditional sketch-based image or video search systems rely on machine learning concepts as their core technology. However, in many applications, machine learning alone is impractical since videos may not be semantically annotated sufficiently, there may be a lack of suitable training data, and the search requirements of the user may frequently change for different tasks. In this work, we develop a visual analytics systems that overcomes the shortcomings of the traditional approach. We make use of a sketch-based interface to enable users to specify search requirement in a flexible manner without depending on semantic annotation. We employ active machine learning to train different analytical models for different types of search requirements. We use visualization to facilitate knowledge discovery at the different stages of visual analytics. This includes visualizing the parameter space of the trained model, visualizing the search space to support interactive browsing, visualizing candidature search results to support rapid interaction for active learning while minimizing watching videos, and visualizing aggregated information of the search results. We demonstrate the system for searching spatiotemporal attributes from sports video to identify key instances of the team and player performance.</abstract><type>Journal Article</type><journal>IEEE Transactions on Visualization and Computer Graphics</journal><volume>19</volume><journalNumber>12</journalNumber><paginationStart>2109</paginationStart><paginationEnd>2118</paginationEnd><publisher/><keywords/><publishedDay>16</publishedDay><publishedMonth>10</publishedMonth><publishedYear>2013</publishedYear><publishedDate>2013-10-16</publishedDate><doi>10.1109/TVCG.2013.207</doi><url/><notes></notes><college>COLLEGE NANME</college><department>Computer Science</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>SCS</DepartmentCode><institution>Swansea University</institution><apcterm/><lastEdited>2019-06-21T13:02:05.1385895</lastEdited><Created>2014-01-09T15:22:04.7619923</Created><path><level id="1">Faculty of Science and Engineering</level><level id="2">School of Mathematics and Computer Science - Computer Science</level></path><authors><author><firstname>Philip A</firstname><surname>Legg</surname><order>1</order></author><author><firstname>David H. S</firstname><surname>Chung</surname><order>2</order></author><author><firstname>Matthew L</firstname><surname>Parry</surname><order>3</order></author><author><firstname>Rhodri</firstname><surname>Bown</surname><order>4</order></author><author><firstname>Mark</firstname><surname>Jones</surname><orcid>0000-0001-8991-1190</orcid><order>5</order></author><author><firstname>Iwan W</firstname><surname>Griffiths</surname><order>6</order></author><author><firstname>Min</firstname><surname>Chen</surname><order>7</order></author></authors><documents><document><filename>0016829-07052015134838.pdf</filename><originalFilename>2013_video__search.pdf</originalFilename><uploaded>2015-05-07T13:48:38.6430000</uploaded><type>Output</type><contentLength>13470909</contentLength><contentType>application/pdf</contentType><version>Version of Record</version><cronfaStatus>true</cronfaStatus><embargoDate>2015-05-07T00:00:00.0000000</embargoDate><documentNotes/><copyrightCorrect>true</copyrightCorrect></document></documents><OutputDurs/></rfc1807> |
spelling |
2019-06-21T13:02:05.1385895 v2 16829 2014-01-09 Transformation of an Uncertain Video Search Pipeline to a Sketch-Based Visual Analytics Loop 2e1030b6e14fc9debd5d5ae7cc335562 0000-0001-8991-1190 Mark Jones Mark Jones true false 2014-01-09 SCS Traditional sketch-based image or video search systems rely on machine learning concepts as their core technology. However, in many applications, machine learning alone is impractical since videos may not be semantically annotated sufficiently, there may be a lack of suitable training data, and the search requirements of the user may frequently change for different tasks. In this work, we develop a visual analytics systems that overcomes the shortcomings of the traditional approach. We make use of a sketch-based interface to enable users to specify search requirement in a flexible manner without depending on semantic annotation. We employ active machine learning to train different analytical models for different types of search requirements. We use visualization to facilitate knowledge discovery at the different stages of visual analytics. This includes visualizing the parameter space of the trained model, visualizing the search space to support interactive browsing, visualizing candidature search results to support rapid interaction for active learning while minimizing watching videos, and visualizing aggregated information of the search results. We demonstrate the system for searching spatiotemporal attributes from sports video to identify key instances of the team and player performance. Journal Article IEEE Transactions on Visualization and Computer Graphics 19 12 2109 2118 16 10 2013 2013-10-16 10.1109/TVCG.2013.207 COLLEGE NANME Computer Science COLLEGE CODE SCS Swansea University 2019-06-21T13:02:05.1385895 2014-01-09T15:22:04.7619923 Faculty of Science and Engineering School of Mathematics and Computer Science - Computer Science Philip A Legg 1 David H. S Chung 2 Matthew L Parry 3 Rhodri Bown 4 Mark Jones 0000-0001-8991-1190 5 Iwan W Griffiths 6 Min Chen 7 0016829-07052015134838.pdf 2013_video__search.pdf 2015-05-07T13:48:38.6430000 Output 13470909 application/pdf Version of Record true 2015-05-07T00:00:00.0000000 true |
title |
Transformation of an Uncertain Video Search Pipeline to a Sketch-Based Visual Analytics Loop |
spellingShingle |
Transformation of an Uncertain Video Search Pipeline to a Sketch-Based Visual Analytics Loop Mark Jones |
title_short |
Transformation of an Uncertain Video Search Pipeline to a Sketch-Based Visual Analytics Loop |
title_full |
Transformation of an Uncertain Video Search Pipeline to a Sketch-Based Visual Analytics Loop |
title_fullStr |
Transformation of an Uncertain Video Search Pipeline to a Sketch-Based Visual Analytics Loop |
title_full_unstemmed |
Transformation of an Uncertain Video Search Pipeline to a Sketch-Based Visual Analytics Loop |
title_sort |
Transformation of an Uncertain Video Search Pipeline to a Sketch-Based Visual Analytics Loop |
author_id_str_mv |
2e1030b6e14fc9debd5d5ae7cc335562 |
author_id_fullname_str_mv |
2e1030b6e14fc9debd5d5ae7cc335562_***_Mark Jones |
author |
Mark Jones |
author2 |
Philip A Legg David H. S Chung Matthew L Parry Rhodri Bown Mark Jones Iwan W Griffiths Min Chen |
format |
Journal article |
container_title |
IEEE Transactions on Visualization and Computer Graphics |
container_volume |
19 |
container_issue |
12 |
container_start_page |
2109 |
publishDate |
2013 |
institution |
Swansea University |
doi_str_mv |
10.1109/TVCG.2013.207 |
college_str |
Faculty of Science and Engineering |
hierarchytype |
|
hierarchy_top_id |
facultyofscienceandengineering |
hierarchy_top_title |
Faculty of Science and Engineering |
hierarchy_parent_id |
facultyofscienceandengineering |
hierarchy_parent_title |
Faculty of Science and Engineering |
department_str |
School of Mathematics and Computer Science - Computer Science{{{_:::_}}}Faculty of Science and Engineering{{{_:::_}}}School of Mathematics and Computer Science - Computer Science |
document_store_str |
1 |
active_str |
0 |
description |
Traditional sketch-based image or video search systems rely on machine learning concepts as their core technology. However, in many applications, machine learning alone is impractical since videos may not be semantically annotated sufficiently, there may be a lack of suitable training data, and the search requirements of the user may frequently change for different tasks. In this work, we develop a visual analytics systems that overcomes the shortcomings of the traditional approach. We make use of a sketch-based interface to enable users to specify search requirement in a flexible manner without depending on semantic annotation. We employ active machine learning to train different analytical models for different types of search requirements. We use visualization to facilitate knowledge discovery at the different stages of visual analytics. This includes visualizing the parameter space of the trained model, visualizing the search space to support interactive browsing, visualizing candidature search results to support rapid interaction for active learning while minimizing watching videos, and visualizing aggregated information of the search results. We demonstrate the system for searching spatiotemporal attributes from sports video to identify key instances of the team and player performance. |
published_date |
2013-10-16T03:19:19Z |
_version_ |
1763750507361337344 |
score |
11.021648 |