A hybrid unsupervised image re-ranking approach with latent topic contents

  • Authors:

  • Lei Zhang;Piji Li;Jun Ma

  • Affiliations:

  • Shandong University, Jinan, China;Shandong University, Jinan, China;Shandong University, Jinan, China

  • Venue:
  • Proceedings of the ACM International Conference on Image and Video Retrieval
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

Although web-scale image search engines such as Google and Bing can provide rough sets of image data, results are noisy and lacking visual diversity. In this paper, our objective is to re-rank initial search results from these image search engines to improve user experience. We present a hybrid approach to capture the benefits of the reciprocal election algorithm proposed by R. van Leuken et al. and the greedy search algorithm proposed by T. Deselaers et al. At first, each image casts votes for other images according to visual similarity. The images with the top highest votes are selected as candidate representatives. Then a bounded greedy selection algorithm is employed to select the most novel and relevant one as the cluster representative. We fuse different visual features to calculate image similarity including color, texture, and especially topic content features. We present an evaluation of pLSA and LDA as dimension reduction approach for the task of web image re-ranking and discuss the benefits of integrating topic distribution features. Extensive experiments demonstrate that using our approach to re-rank an initially returned set of images from Google and Bing search engines is a practical way to improve the user satisfaction in terms of cluster recall, F1 score and the harmonic mean of NDCG and cluster recall.