Improving retrieval performance by region constraints and relevance feedback

  • Authors:
  • Tao Wang;Young Rui;Jia-Guang Sun

  • Affiliations:
  • Department of Computer Science and Technology, Tsinghua University, Beijing 100084, P.R. China and Statistical Computing Group, Intel China Research Center, Beijing 100020, P.R. China;Microsoft Research, One Microsoft Way, Redmond, WA;Department of Computer Science and Technology, Tsinghua University, Beijing 100084, P.R. China

  • Venue:
  • Journal of Computer Science and Technology
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this paper, region features and relevance feedback are used to improve the performance of CBIR. Unlike existing region-based approaches where either individual regions are used or only simple spatial layout is modeled, the proposed approach simultaneously models both region properties and their spatial relationships in a probabilistic framework. Furthermore, the retrieval performance is improved by an adaptive filter based relevance feedback. To illustrate the performance of the proposed approach, extensive experiments have been carried out on a large heterogeneous image collection with 17,000 images, which render promising results on a wide variety of queries.