The Open UniversitySkip to content

Detecting regions of interest using eye tracking for CBIR

Ren, Qingtao; Chen, Yongqiang; Zhang, Peng; Song, Dawei and Hou, Yuexian (2015). Detecting regions of interest using eye tracking for CBIR. In: SIGIR2015 Workshop on Neuro-Physiological Methods in IR Research, 13 Aug 2015, Santiago, Chile.

Full text available as:
PDF (Version of Record) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (213kB) | Preview
Google Scholar: Look up in Google Scholar


Identifying Regions of Interest (ROIs) in images has been shown an effective way to enhance the performance of Content Based Image Retrieval (CBIR). Most existing ROI identification methods are based on salience detection, and the identified ROIs may not be the regions that users are really interested in. While manual selection of ROIs can directly reflect users’ interests, it puts extra cognitive overhead to users. To alleviate these limitations, in this paper, we propose a novel eye-tracking based method to detect ROIs for CBIR, in an unobtrusive way. Experimental results have demonstrated that our model performed effectively compared with various state of the art methods.

Item Type: Conference or Workshop Item
Copyright Holders: 2015 The Authors
Project Funding Details:
Funded Project NameProject IDFunding Body
Not Set2013CB329304Chinese 973 Program
Not Set2014CB744604Chinese 973 Program
Not Set2015AA015403Chinese 863 Program
Not Set61272265Natural Science Foundation of China
Not Set61402324Natural Science Foundation of China
Keywords: content-based image retrieval; eye tracking; regions of interest
Academic Unit/School: Faculty of Science, Technology, Engineering and Mathematics (STEM) > Computing and Communications
Faculty of Science, Technology, Engineering and Mathematics (STEM)
Related URLs:
Item ID: 44134
Depositing User: Dawei Song
Date Deposited: 24 Aug 2015 10:02
Last Modified: 07 Dec 2018 22:57
Share this page:

Download history for this item

These details should be considered as only a guide to the number of downloads performed manually. Algorithmic methods have been applied in an attempt to remove automated downloads from the displayed statistics but no guarantee can be made as to the accuracy of the figures.

Actions (login may be required)

Policies | Disclaimer

© The Open University   contact the OU