COPE: interactive image retrieval using conversational recommendation

Bartoz Balcer, Martin Halvey, Stephen Brewster, Joemon Jose

Research output: Contribution to conferencePaper

Abstract

Most multimedia retrieval services e.g. YouTube, Flickr, Google etc. rely on users searching using textual queries or examples. However, this solution is inadequate when there is no text, very little text, the text is in a foreign language or the user cannot form textual a query. In order to overcome these shortcomings we have developed an image retrieval system called COPE (COnversational Picture Exploration) that can use a number of different preference feedback mechanisms, inspired by conversational recommendation paradigms, for image retrieval. In COPE users are presented with a small number of search results and simply have to express whether these results match their information need. We examine the suitability of a number of feedback approaches for semiautomatic and interactive image retrieval. For interactive retrieval we compared our preference based approaches to text based search (where we consider text to be an upper bound), our results indicate that users prefer preference based search to text based search and in some cases our approaches can outperform text based search.

Conference

Conference26th Annual BCS Interaction Specialist Group Conference on People and Computers
Abbreviated titleBCS-HCI 2012
CountryUnited Kingdom
CityBirmingham
Period10/09/12 → …

Fingerprint

Image retrieval
Feedback

Keywords

  • multimedia retrieval
  • image retrieval
  • text based searching

Cite this

Balcer, B., Halvey, M., Brewster, S., & Jose, J. (2012). COPE: interactive image retrieval using conversational recommendation. 1-10. Paper presented at 26th Annual BCS Interaction Specialist Group Conference on People and Computers , Birmingham, United Kingdom.
Balcer, Bartoz ; Halvey, Martin ; Brewster, Stephen ; Jose, Joemon. / COPE : interactive image retrieval using conversational recommendation. Paper presented at 26th Annual BCS Interaction Specialist Group Conference on People and Computers , Birmingham, United Kingdom.
@conference{28cfe35cf732494c857f7b0a80d3833a,
title = "COPE: interactive image retrieval using conversational recommendation",
abstract = "Most multimedia retrieval services e.g. YouTube, Flickr, Google etc. rely on users searching using textual queries or examples. However, this solution is inadequate when there is no text, very little text, the text is in a foreign language or the user cannot form textual a query. In order to overcome these shortcomings we have developed an image retrieval system called COPE (COnversational Picture Exploration) that can use a number of different preference feedback mechanisms, inspired by conversational recommendation paradigms, for image retrieval. In COPE users are presented with a small number of search results and simply have to express whether these results match their information need. We examine the suitability of a number of feedback approaches for semiautomatic and interactive image retrieval. For interactive retrieval we compared our preference based approaches to text based search (where we consider text to be an upper bound), our results indicate that users prefer preference based search to text based search and in some cases our approaches can outperform text based search.",
keywords = "multimedia retrieval , image retrieval, text based searching",
author = "Bartoz Balcer and Martin Halvey and Stephen Brewster and Joemon Jose",
year = "2012",
month = "9",
day = "10",
language = "English",
pages = "1--10",
note = "26th Annual BCS Interaction Specialist Group Conference on People and Computers , BCS-HCI 2012 ; Conference date: 10-09-2012",

}

Balcer, B, Halvey, M, Brewster, S & Jose, J 2012, 'COPE: interactive image retrieval using conversational recommendation' Paper presented at 26th Annual BCS Interaction Specialist Group Conference on People and Computers , Birmingham, United Kingdom, 10/09/12, pp. 1-10.

COPE : interactive image retrieval using conversational recommendation. / Balcer, Bartoz; Halvey, Martin; Brewster, Stephen; Jose, Joemon.

2012. 1-10 Paper presented at 26th Annual BCS Interaction Specialist Group Conference on People and Computers , Birmingham, United Kingdom.

Research output: Contribution to conferencePaper

TY - CONF

T1 - COPE

T2 - interactive image retrieval using conversational recommendation

AU - Balcer, Bartoz

AU - Halvey, Martin

AU - Brewster, Stephen

AU - Jose, Joemon

PY - 2012/9/10

Y1 - 2012/9/10

N2 - Most multimedia retrieval services e.g. YouTube, Flickr, Google etc. rely on users searching using textual queries or examples. However, this solution is inadequate when there is no text, very little text, the text is in a foreign language or the user cannot form textual a query. In order to overcome these shortcomings we have developed an image retrieval system called COPE (COnversational Picture Exploration) that can use a number of different preference feedback mechanisms, inspired by conversational recommendation paradigms, for image retrieval. In COPE users are presented with a small number of search results and simply have to express whether these results match their information need. We examine the suitability of a number of feedback approaches for semiautomatic and interactive image retrieval. For interactive retrieval we compared our preference based approaches to text based search (where we consider text to be an upper bound), our results indicate that users prefer preference based search to text based search and in some cases our approaches can outperform text based search.

AB - Most multimedia retrieval services e.g. YouTube, Flickr, Google etc. rely on users searching using textual queries or examples. However, this solution is inadequate when there is no text, very little text, the text is in a foreign language or the user cannot form textual a query. In order to overcome these shortcomings we have developed an image retrieval system called COPE (COnversational Picture Exploration) that can use a number of different preference feedback mechanisms, inspired by conversational recommendation paradigms, for image retrieval. In COPE users are presented with a small number of search results and simply have to express whether these results match their information need. We examine the suitability of a number of feedback approaches for semiautomatic and interactive image retrieval. For interactive retrieval we compared our preference based approaches to text based search (where we consider text to be an upper bound), our results indicate that users prefer preference based search to text based search and in some cases our approaches can outperform text based search.

KW - multimedia retrieval

KW - image retrieval

KW - text based searching

UR - http://dl.acm.org/citation.cfm?id=2377918&CFID=981374015&CFTOKEN=43955908

M3 - Paper

SP - 1

EP - 10

ER -

Balcer B, Halvey M, Brewster S, Jose J. COPE: interactive image retrieval using conversational recommendation. 2012. Paper presented at 26th Annual BCS Interaction Specialist Group Conference on People and Computers , Birmingham, United Kingdom.