The Bayesian image retrieval system, PicHunter: theory, implementation, and psychophysical experiments

@article{citeulike:557224, abstract = {Presents the theory, design principles, implementation and performance results of PicHunter, a prototype content-based image retrieval (CBIR) system. In addition, this document presents the rationale, design and results of psychophysical experiments that were conducted to address some key issues that arose during PicHunter's development. The PicHunter project makes four primary contributions to research on CBIR. First, PicHunter represents a simple instance of a general Bayesian framework which we describe for using relevance feedback to direct a search. With an explicit model of what users would do, given the target image they want, PicHunter uses Bayes's rule to predict the target they want, given their actions. This is done via a probability distribution over possible image targets, rather than by refining a query. Second, an entropy-minimizing display algorithm is described that attempts to maximize the information obtained from a user at each iteration of the search. Third, PicHunter makes use of hidden annotation rather than a possibly inaccurate/inconsistent annotation structure that the user must learn and make queries in. Finally, PicHunter introduces two experimental paradigms to quantitatively evaluate the performance of the system, and psychophysical experiments are presented that support the theoretical claims}, author = {Cox, I. J. and Miller, M. L. and Minka, T. P. and Papathomas, T. V. and Yianilos, P. N. }, citeulike-article-id = {557224}, doi = {10.1109/83.817596}, journal = {Image Processing, IEEE Transactions on}, keywords = {cbir}, number = {1}, pages = {20--37}, posted-at = {2008-04-13 22:26:46}, priority = {1}, title = {The Bayesian image retrieval system, PicHunter: theory, implementation, and psychophysical experiments}, url = {http://dx.doi.org/10.1109/83.817596}, volume = {9}, year = {2000} }

TY - JOUR ID - citeulike:557224 L3 - citeulike-article-id:557224 TI - The Bayesian image retrieval system, PicHunter: theory, implementation, and psychophysical experiments JF - Image Processing, IEEE Transactions on VL - 9 IS - 1 SP - 20 EP - 37 N2 - Presents the theory, design principles, implementation and performance results of PicHunter, a prototype content-based image retrieval (CBIR) system. In addition, this document presents the rationale, design and results of psychophysical experiments that were conducted to address some key issues that arose during PicHunter's development. The PicHunter project makes four primary contributions to research on CBIR. First, PicHunter represents a simple instance of a general Bayesian framework which we describe for using relevance feedback to direct a search. With an explicit model of what users would do, given the target image they want, PicHunter uses Bayes's rule to predict the target they want, given their actions. This is done via a probability distribution over possible image targets, rather than by refining a query. Second, an entropy-minimizing display algorithm is described that attempts to maximize the information obtained from a user at each iteration of the search. Third, PicHunter makes use of hidden annotation rather than a possibly inaccurate/inconsistent annotation structure that the user must learn and make queries in. Finally, PicHunter introduces two experimental paradigms to quantitatively evaluate the performance of the system, and psychophysical experiments are presented that support the theoretical claims KW - cbir AU - Cox, IJ AU - Miller, ML AU - Minka, TP AU - Papathomas, TV AU - Yianilos, PN PY - 2000/// UR - http://dx.doi.org/10.1109/83.817596 ER -