This paper presents an object recognition method based on recursive neural networks (RNNs) and multiresolution trees (MRTs). MRTs are a novel hierarchical structure proposed to represent both the set of homogeneous regions in which images can be divided and the evolution of the segmentation process performed to determine such regions. Moreover, knowing the optimal number of regions that should be extracted from the images is not critical for the construction of MRTs, that are also invariant w.r.t. rotations and translations. A set of experiments was performed on a subset of the Caltech benchmark database, comparing the performances of the MRT and directed acyclic graph (DAG) representations. The results obtained by the proposed object detection technique are also very promising in comparison with other state-of-the-art approaches available in the literature.


Object Recognition Directed Acyclic Graph Homogeneous Region Recurrent Neural Network Label Edge 
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.
    Frasconi, P., Gori, M., Sperduti, A.: A general framework for adaptive processing of data structures. IEEE Transactions on Neural Networks 9, 768–786 (1998)CrossRefMATHGoogle Scholar
  2. 2.
    Bianchini, M., Maggini, M., Sarti, L., Scarselli, F.: Recursive neural networks learn to localize faces. Pattern Recognition Letters, 1885–1895 (2005)Google Scholar
  3. 3.
    Bianchini, M., Maggini, M., Sarti, L., Scarselli, F.: Recursive neural networks for object detection. In: Proceedings of IEEE IJCNN, pp. 1911–1915 (2004)Google Scholar
  4. 4.
    Hunter, G.M., Steiglitz, K.: Operations on images using quadtrees. IEEE Transactions PAMI 1, 145–153 (1979)Google Scholar
  5. 5.
    Song, Y., Zhang, A.: Monotonic tree. In: Proceedings of the 10th Intl. Conf. on Discrete Geometry for Computer Imagery, Bordeaux – France (2002)Google Scholar
  6. 6.
    Roubal, J., Peucker, T.: Automated contour labeling and the contour tree. Proceedings of AUTO-CARTO 7, 472–481 (1985)Google Scholar
  7. 7.
    Fergus, R., Perona, P., Zisserman, A.: Object class recognition by unsupervised scale–invariant learning. In: Proceedings of IEEE CVPR, pp. 264–271 (2003)Google Scholar
  8. 8.
    Zhang, W., Yu, B., Zelinsky, G., Samaras, D.: Object class recognition using multiple layer boosting with heterogeneous features. In: Proceedings of CVPR, vol. 2, pp. 323–330 (2005)Google Scholar
  9. 9.
    Opelt, A., Fussenegger, M., Pinz, A., Auer, A.: Weak Hypotheses and Boosting for Generic Object Detection and Recognition. In: Pajdla, T., Matas, J(G.) (eds.) ECCV 2004. LNCS, vol. 3022, pp. 71–84. Springer, Heidelberg (2004)CrossRefGoogle Scholar
  10. 10.
    Thureson, J., Carlsson, S.: Appearance Based Qualitative Image Description for Object Class Recognition. In: Pajdla, T., Matas, J(G.) (eds.) ECCV 2004. LNCS, vol. 3022, pp. 518–529. Springer, Heidelberg (2004)CrossRefGoogle Scholar
  11. 11.
    Küchler, A., Goller, C.: Inductive learning in symbolic domains using structure–driven recurrent neural networks. In: Görz, G., Hölldobler, S. (eds.) Advances in Artificial Intelligence, pp. 183–197. Springer, Berlin (1996)Google Scholar
  12. 12.
    Bianchini, M., Maggini, M., Sarti, L., Scarselli, F.: Recursive neural networks for processing graphs with labelled edges: Theory and applications. Neural Networks, 1040–1050 (2005)Google Scholar
  13. 13.
    de Mauro, C., Diligenti, M., Gori, M., Maggini, M.: Similarity learning for graph based image representation. Pattern Recognition Letters 24, 1115–1122 (2003)MATHCrossRefGoogle Scholar
  14. 14.
    Gori, M., Maggini, M., Sarti, L.: A recursive neural network model for processing directed acyclic graphs with labeled edges. In: Proceedings of IEEE IJCNN, pp. 1351–1355 (2003)Google Scholar
  15. 15.
    Bengio, Y., Frasconi, P., Simard, P.: Learning long–term dependencies with gradient descent is difficult. IEEE Transactions on Neural Networks 5, 157–166 (1994)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Monica Bianchini
    • 1
  • Marco Maggini
    • 1
  • Lorenzo Sarti
    • 1
  1. 1.DIIUniversità degli Studi di SienaSienaItaly

Personalised recommendations