Uniqueness Filtering for Local Feature Descriptors in Urban Building Recognition

  • Giang Phuong Nguyen
  • Hans Jørgen Andersen
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5099)


Existing local feature detectors such as Scale Invariant Feature Transform (SIFT) usually produce a large number of features per image. This is a major disadvantage in terms of the speed of search and recognition in a run-time application. Besides, not all detected features are equally important in the search. It is therefore essential to select informative descriptors. In this paper, we propose a new approach to selecting a subset of local feature descriptors. Uniqueness is used as a filtering criterion in selecting informative features. We formalize the notion of uniqueness and show how it can be used for selection purposes. To evaluate our approach, we carried out experiments in urban building recognition domains with different datasets. The results show a significant improvement not only in recognition speed, as a result of using fewer features, but also in the performance of the system with selected features.


Feature Vector Recognition Rate Interest Point Informative Descriptor Vocabulary Tree 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


  1. 1.
    Mikolajczyk, K., Schmid, C.: A performance evaluation of local descriptors. IEEE Transactions on Pattern Analysis & Machine Intelligence 27(10), 1615–1630 (2005)CrossRefGoogle Scholar
  2. 2.
    Lowe, D.G.: Distinctive image features from scale-invariant keypoints. International Journal of Computer Vision 60(2), 91–110 (2004)CrossRefGoogle Scholar
  3. 3.
    Ke, Y., Sukthankar, R.: Pca-sift: A more distinctive representation for local image descriptors. In: IEEE Conf. on Computer Vision and Pattern Recognition, vol. 2, pp. 506–513 (2004)Google Scholar
  4. 4.
    Brown, M., Szeliski, R., Winder, S.: Multi-image matching using multi-scale oriented patches. In: IEEE Conf. on Computer Vision and Pattern Recognition, vol. 1, pp. 510–517 (2005)Google Scholar
  5. 5.
    Nguyen, G.P., Andersen, H., Christensen, M.: Urban building recognition during significant temporal variations. In: IEEE Workshop on Application of Computer Vision (2008)Google Scholar
  6. 6.
    Royer, E., et al.: Localization in urban environments: monocular vision compared to a differential gps sensor. In: IEEE Conf. on Computer Vision and Pattern Recognition, vol. 2, pp. 114–121 (2005)Google Scholar
  7. 7.
    Kosecka, J., Li, F., Yang, X.: Global localization and relative positioning based on scale-invariant keypoints. Robotics and Autonomous Systems 52(1), 27–38 (2005)CrossRefGoogle Scholar
  8. 8.
    Robertson, D., Cipolla, R.: An image based system for urban navigation. In: British Machine Vision Conference (2004)Google Scholar
  9. 9.
    Nister, D., Stewenius, H.: Scalable recognition with a vocabulary tree. In: IEEE Conf. on Computer Vision and Pattern Recognition, vol. 2, pp. 2161–2168 (2006)Google Scholar
  10. 10.
    Li, F., Kosecka, J.: Probabilistic location recognition using reduced feature set. In: IEEE Intl. Conf. on Robotics and Automation, pp. 3405–3410 (2006)Google Scholar
  11. 11.
    Fritz, G., Seifert, C., Paletta, L.: Urban object recognition from informative local features. In: IEEE Intl. Conf. on Robotics and Automation, pp. 132–138 (2005)Google Scholar
  12. 12.
    Kadir, T., Brady, M.: Saliency, scale and image description. International Journal of Computer Vision 2(45), 83–105 (2001)CrossRefGoogle Scholar
  13. 13.
    Schiele, B., Crowley, J.L.: Probabilistic object recognition using multidimensional receptive field histograms. In: International Conference on Pattern Recognition (1996)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2008

Authors and Affiliations

  • Giang Phuong Nguyen
    • 1
  • Hans Jørgen Andersen
    • 1
  1. 1.Department of Media Technology and Engineering ScienceAalborg UniversityDenmark

Personalised recommendations