Discriminative features for image classification and retrieval

In this paper, we present a new method to improve the performance of current bag-of-word based image classification process. After feature extraction, we introduce a pairwise image matching scheme to select the discriminative features. Only the labeled information from the training-sets is used to update the feature weights via an iterative matching processing. The selected features correspond to the foreground content of the images thus highlight the high level category knowledge of images. "Visual words" are constructed on these selected features. Our method can be used as a refinement step for current image classification and retrieval process. We prove the efficiency of our method in three tasks: supervised image classification, semi-supervised image classification and image retrieval. In the experimental part, two canonical datasets Caltech 256 and MSRC-v2 are used. Our methods have increased the performance of given image analysis tasks. The accuracies of supervised and semi-supervised image classification has been increased up to 21%. Meanwhile, the precision of image retrieval results has also been improved by using our method.