Learning to Rank Images from Eye Movements
Learning to Rank Images from Eye Movements
Combining multiple information sources can improve the accuracy of search in information retrieval. This paper presents a new image search strategy which combines image features together with implicit feedback from users' eye movements, using them to rank images. In order to better deal with larger data sets, we present a perceptron formulation of the Ranking Support Vector Machine algorithm. We present initial results on inferring the rank of images presented in a page based on simple image features and implicit feedback of users. The results show that the perceptron algorithm improves the results, and that fusing eye movements and image histograms gives better rankings to images than either of these features alone.
2009-2016
Pasupa, Kitsuchart
952ededb-8c97-41b7-a65b-6aba31de2669
Saunders, Craig
26634635-4d4d-4469-b9ec-1d68788aa47a
Szedmak, Sandor
c6a84aa3-2956-4acf-8293-a1b676f6d7d8
Klami, Arto
7d24333d-4417-42ba-a188-c22dea625f1d
Kaski, Samuel
87dad6b5-fda3-494f-84f9-1de967bf1458
Gunn, Steve
306af9b3-a7fa-4381-baf9-5d6a6ec89868
4 October 2009
Pasupa, Kitsuchart
952ededb-8c97-41b7-a65b-6aba31de2669
Saunders, Craig
26634635-4d4d-4469-b9ec-1d68788aa47a
Szedmak, Sandor
c6a84aa3-2956-4acf-8293-a1b676f6d7d8
Klami, Arto
7d24333d-4417-42ba-a188-c22dea625f1d
Kaski, Samuel
87dad6b5-fda3-494f-84f9-1de967bf1458
Gunn, Steve
306af9b3-a7fa-4381-baf9-5d6a6ec89868
Pasupa, Kitsuchart, Saunders, Craig, Szedmak, Sandor, Klami, Arto, Kaski, Samuel and Gunn, Steve
(2009)
Learning to Rank Images from Eye Movements.
Proceeding of 2009 IEEE 12th International Conference on Computer Vision (ICCV'2009) Workshop on Human-Computer Interaction (HCI'2009), Kyoto, Japan.
27 Sep - 04 Oct 2009.
.
Record type:
Conference or Workshop Item
(Other)
Abstract
Combining multiple information sources can improve the accuracy of search in information retrieval. This paper presents a new image search strategy which combines image features together with implicit feedback from users' eye movements, using them to rank images. In order to better deal with larger data sets, we present a perceptron formulation of the Ranking Support Vector Machine algorithm. We present initial results on inferring the rank of images presented in a page based on simple image features and implicit feedback of users. The results show that the perceptron algorithm improves the results, and that fusing eye movements and image histograms gives better rankings to images than either of these features alone.
Text
learning_to_rank_images_from_eye_movements.pdf
- Version of Record
More information
Published date: 4 October 2009
Additional Information:
Event Dates: 27 September - 4 October 2009
Venue - Dates:
Proceeding of 2009 IEEE 12th International Conference on Computer Vision (ICCV'2009) Workshop on Human-Computer Interaction (HCI'2009), Kyoto, Japan, 2009-09-27 - 2009-10-04
Organisations:
Electronic & Software Systems
Identifiers
Local EPrints ID: 267964
URI: http://eprints.soton.ac.uk/id/eprint/267964
PURE UUID: 09b8bf96-eb94-42e9-ab91-14bc11c24e88
Catalogue record
Date deposited: 27 Sep 2009 14:54
Last modified: 14 Mar 2024 09:01
Export record
Contributors
Author:
Kitsuchart Pasupa
Author:
Craig Saunders
Author:
Sandor Szedmak
Author:
Arto Klami
Author:
Samuel Kaski
Author:
Steve Gunn
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics