Learning Distance Metrics with Feature Space Performance for Image Retrieval

Conference paper
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 288)


Learning from samples in cases where many high-dimensional vectors but only few samples are available is commonly considered a challenging problem in content-based image retrieval (CBIR). In this paper, we propose an algorithm for metric learning based on spatial distribution of image features. The optimal distance metric is then found by minimizing the divergence between the two distributions. The key idea is to construct a global metric matrix that minimizes the cluster distortions, namely, one that reduces high variances and expands low variances for the data to make a spherical form as good as possible in the high-dimensional data spaces. Experimental results show that our approach is effective in improving the performance of CBIR systems.


CBIR Image feature space Cluster geometry Learning distance metric 



This research was partially supported by “the Fundamental Research Funds for the Central Universities (No. 13D11205).”


  1. 1.
    Hoi SCH, Liu W, Lyu MR, Ma W-Y (2006) Learning distance metrics with contextual constraints for image retrieval. In: Proceedings of the computer vision and pattern recognitionGoogle Scholar
  2. 2.
    Shental N, Hertz T, Weinshall D, Pavel M (2002) Adjustment learning and relevant component analysis. In: ECCV 2002, pp 776–792Google Scholar
  3. 3.
    Davis JV, Kulis B, Jain P, Sra S, Dhillon IS (2007) Information theoretic metric learning. In: Proceedings of the international conference on machine learning, Corvalis, Oregon, pp 209–216Google Scholar
  4. 4.
    Yang L, Sukthankar R, Hoi SCH (2010) A boosting framework for visuality-preserving distance metric learning and its application to medical image retrieval. IEEE Trans Pattern Anal Mach Intell 32(1):30–44Google Scholar
  5. 5.
    Chen J, Wang R, Shan S, Chen X, Gao X (2006) Isomap based on the image euclidean distance. In: The IEEE 7th international conference on pattern recognition (ICPR2006), pp 1110–1113Google Scholar
  6. 6.
    Wang Liwei, Zhang Yan, Feng Jufu (2005) On the Euclidean distance of images. IEEE Trans Pattern Anal Mach Intell 27(8):1334–1339CrossRefGoogle Scholar
  7. 7.
    Luo X, Shishibori M, Ren F, Kita K (2007) Incorporate feature space transformation to content-based image retrieval with relevance feedback. Int J Innovative Comput Inf Control (IJICIC) 3(5):1237–1250Google Scholar
  8. 8.
    Balmachnova E, Florack L, ter Haar Romeny B (2007) Feature vector similarity based on local structure. In: SSVM 2007, pp 386–393Google Scholar
  9. 9.
    Ishikawa Y, Subramanya R, Faloutsos C (1998) MindReader: querying database through multiple examples. In: Proceedings of the 24th international conference on very large database, pp 218–227Google Scholar
  10. 10.
    Niblack W, Barber R, Equitz W, Flickner M, Glasman E, Pektovic D, Yanker P, Faloutsos C, Taubin G (1993) The QBIC project: querying images by content using color, texture, and shape. In: Proceedings of SPIE storage and retrieval for image and video databases, pp 173–181Google Scholar
  11. 11.
    Manjunath BS, Ma WY (1996) Texture features for browsing and retrieval of large image data. IEEE Trans PAMI 18(8):837–842CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2014

Authors and Affiliations

  1. 1.School of Computer Science and TechnologyDonghua UniversitySongjiang DistrictChina
  2. 2.Faculty and School of EngineeringThe University of TokushimaTokushimaJapan

Personalised recommendations