A unifying view of image similarity

We study solutions to the problem of evaluating image similarity in the context of content-based image retrieval (CBIR). Retrieval is formulated as a classification problem, where the goal is to minimize probability of retrieval error. It is shown that this formulation establishes a common ground for comparing similarity functions, exposes assumptions hidden behind in most commonly used ones, enables a critical analysis of their relative merits, and determines the retrieval scenarios for which each may be most suited. We conclude that most of the current similarity functions are sub-optimal special cases of the Bayesian criteria that results from explicit minimization of error probability.

[1]  Anil K. Jain,et al.  Texture classification and segmentation using multiresolution simultaneous autoregressive models , 1992, Pattern Recognit..

[2]  Thomas M. Cover,et al.  Elements of Information Theory , 2005 .

[3]  Keinosuke Fukunaga,et al.  Introduction to Statistical Pattern Recognition , 1972 .

[4]  P. Meer,et al.  Retrieval performance improvement through low rank corrections , 1999, Proceedings IEEE Workshop on Content-Based Access of Image and Video Libraries (CBAIVL'99).

[5]  Bernt Schiele,et al.  Probabilistic object recognition using multidimensional receptive field histograms , 1996, Proceedings of 13th International Conference on Pattern Recognition.

[6]  Edward Courtney,et al.  5 = 10 M , 1993 .

[7]  Henning Biermann,et al.  Defining image content with multiple regions-of-interest , 1999, Proceedings IEEE Workshop on Content-Based Access of Image and Video Libraries (CBAIVL'99).

[8]  Tom Minka,et al.  Interactive learning with a "society of models" , 1997, Pattern Recognit..

[9]  John W. Fisher,et al.  Flexible histograms: a multiresolution target discrimination model , 1998, Defense, Security, and Sensing.

[10]  Rosalind W. Picard,et al.  Interactive Learning Using a "Society of Models" , 2017, CVPR 1996.

[11]  Bernt Schiele,et al.  Object Recognition Using Multidimensional Receptive Field Histograms , 1996, ECCV.

[12]  Christos Faloutsos,et al.  QBIC project: querying images by content, using color, texture, and shape , 1993, Electronic Imaging.