A survey on image retrieval performance of different bag of visual words indexing techniques

Abstract

In this paper a survey has been carried out over image retrieval performances of bag of visual words (BoVW) method using different indexing techniques. Bag of visual word method is a content based image retrieval technique, where images are represented as a sparse vector of occurrences of visual words. In this paper different indexing techniques are used to compute near similar visual word vectors of a query image. Locality sensitive hashing, SR-tree based indexing and naive L1 and L2 norm based distance metric calculation are used here. Standard datasets like, UKBench [19], holiday dataset [9] and images from SMARAK1 are used for performance analysis.