PicHunter: Bayesian relevance feedback for image retrieval

• Published in: Proceedings of 13th International Conference on Pattern Recognition Vol. 3; pp. 361 - 369 vol.3
• Main Authors: Cox, I.J., Miller, M.L., Omohundro, S.M., Yianilos, P.N.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 1996
• Subjects: Bayesian methods; Displays; Feedback; History; Image databases; Image retrieval; Information retrieval; Predictive models; Spatial databases; User interfaces
• ISBN: 9780818672828; 081867282X
• ISSN: 1051-4651
• DOI: 10.1109/ICPR.1996.546971

This paper describes PicHunter, an image retrieval system that implements a novel approach to relevance feedback, such that the entire history of user selections contributes to the system's estimate of the user's goal image. To accomplish this, PicHunter uses Bayesian learning based on a probabilistic model of a user's behavior. The predictions of this model are combined with the selections made during a search to estimate the probability associated with each image. These probabilities are then used to select images for display. Details of our model of a user's behavior were tuned using an off-line leaning algorithm. For clarity, our studies were done with the simplest possible user interface but the algorithm can easily be incorporated into systems which support complex queries, including most previously proposed systems. However, even with this constraint and simple image features, PicHunter is able to locate randomly selected targets in a database of 4522 images after displaying an average of only 55 groups of 4 images which is over 10 times better than chance. We therefore expect that the performance of current image database retrieval systems can be improved by incorporation of the techniques described here.