Tripartite Graph Models for Multi Modal Image Retrieval

Most of the traditional image retrieval methods use either low level visual features or embedded text for representation and indexing. In recent years, there has been significant interest in combining these two different modalities for effective retrieval. In this paper, we propose a tri-partite graph based representation of the multi model data for image retrieval tasks. Our representation is ideally suited for dynamically changing or evolving data sets, where repeated semantic indexing is practically impossible. An undirected tripartite graph G = (T,V,D,E) has three sets of vertices where, T = {t1, t2 . . . , tn} are text words, V = {v1,v2 . . . ,vm} are visual words and D = {d1,d2 . . . ,di} are images with E = {ed1 t1 , ..,e di tn ,e d1 v1 , ..,e di vm , et1 v1 , ..,e tn vm} as set of edges. Figure 1 pictorially represent the tripartite graph model (TGM) we use.