Title
Congruency-Based Reranking
Abstract
We present a tool for re-ranking the results of a specific query by considering the (n+1) × (n+1) matrix of pairwise similarities among the elements of the set of n retrieved results and the query itself. The re-ranking thus makes use of the similarities between the various results and does not employ additional sources of information. The tool is based on graphical Bayesian models, which reinforce retrieved items strongly linked to other retrievals, and on repeated clustering to measure the stability of the obtained associations. The utility of the tool is demonstrated within the context of visual search of documents from the Cairo Genizah and for retrieval of paintings by the same artist and in the same style.
Year
DOI
Venue
2014
10.1109/CVPR.2014.270
CVPR
Keywords
Field
DocType
Bayes methods,image retrieval,matrix algebra,Cairo Genizah manuscript,congruency-based reranking,graphical Bayesian model,paintings retrieval,pairwise similarity,visual search,clustering,genizah,graphical models,image retrieval,query retrieval,reranking
Visual search,Pairwise comparison,Information retrieval,Pattern recognition,Query expansion,Computer science,Image retrieval,Artificial intelligence,Graphical model,Cluster analysis,Visual Word,Bayesian probability
Conference
Citations 
PageRank 
References 
3
0.38
4
Authors
9
Name
Order
Citations
PageRank
Itai Ben-Shalom130.72
Noga Levy2564.96
Lior Wolf35501352.38
Nachum Dershowitz432.07
Adiel Ben-Shalom561.98
Roni Shweka6303.96
Yaacov Choueka7241202.83
Tamir Hazan831.05
Yaniv Bar930.72