Multi-net System Configuration for Visual Object Segmentation by Error Backpropagation

  • Ignazio Gallo
  • Marco Vanetti
  • Simone Albertini
  • Angelo Nodari
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7887)

Abstract

This work proposes a new error backpropagation approach as a systematic way to configure and train the Multi-net System MNOD, a recently proposed algorithm able to segment a class of visual objects from real images. First, a single node of the MNOD is configured in order to best resolve the visual object segmentation problem using the best combination of parameters and features. The problem is then how to add new nodes in order to improve accuracy and avoid overfitting situations. In this scenario, the proposed approach employs backpropagation of error maps to add new nodes with the aim of increasing the overall segmentation performance. Experiments conducted on a standard dataset of real images show that our configuration method, using only simple edges and colors descriptors, leads to configurations that produced comparable results in visual objects segmentation.

Keywords

multiple neural networks configuration visual object segmentation 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Al-ani, A.: Ant colony optimization for feature subset selection. In: Enformatika Conferences, pp. 35–38 (2005)Google Scholar
  2. 2.
    Ash, T.: Dynamic node creation in backpropagation networks. Connection Science 1, 365–375 (1989)CrossRefGoogle Scholar
  3. 3.
    Carreira, J., Sminchisescu, C.: Constrained Parametric Min-Cuts for Automatic Object Segmentation. In: IEEE International Conference on Computer Vision and Pattern Recognition (June 2010); Description of our winning PASCAL VOC 2009 and 2010 segmentation entryGoogle Scholar
  4. 4.
    Dalal, N., Triggs, B.: Histograms of oriented gradients for human detection. In: Proc. CVPR, pp. 886–893 (2005)Google Scholar
  5. 5.
    Everingham, M., Van Gool, L., Williams, C.K.I., Winn, J., Zisserman, A.: The PASCAL Visual Object Classes Challenge, VOC 2007-2011 Results, http://www.pascal-network.org/challenges/VOC
  6. 6.
    Freund, Y., Schapire, R.E.: A decision-theoretic generalization of on-line learning and an application to boosting. Journal of Computer and System Sciences 55(1), 119–139 (1997)MathSciNetMATHCrossRefGoogle Scholar
  7. 7.
    Gallo, I., Nodari, A.: Learning object detection using multiple neural netwoks. In: VISAPP 2011. INSTICC Press (2011)Google Scholar
  8. 8.
    Jaccard, P.: The distribution of the flora in the alpine zone. New Phytologist 11(2), 37–50 (1912)CrossRefGoogle Scholar
  9. 9.
    Liu, J., Jia, Y.: Dynamic Construction of Multilayer Neural Networks for Classification (2011)Google Scholar
  10. 10.
    Liu, X., Wang, D.L.: Texture classification using spectral histograms. IEEE Transactions on Image Processing 12, 661–670 (2003)CrossRefGoogle Scholar
  11. 11.
    Ma, Y., Derksen, H., Hong, W., Wright, J.: Segmentation of multivariate mixed data via lossy data coding and compression. IEEE Transactions on Pattern Analysis and Machine Intelligence 29, 1546–1562 (2007)CrossRefGoogle Scholar
  12. 12.
    Nascimento, D.S.C., Canuto, A.M.P., Silva, L.M.M., Coelho, A.L.V.: Combining different ways to generate diversity in bagging models: An evolutionary approach. In: International Symposium on Neural Networks, pp. 2235–2242 (2011)Google Scholar
  13. 13.
    Schapire, R.E.: The strength of weak learnability. Mach. Learn. 5(2), 197–227 (1990)Google Scholar
  14. 14.
    Amanda, J.C.: Multi-Net Systems. In: Combining Artificial Neural Nets: Ensemble and Modular Multi-Net Systems. Springer (1999)Google Scholar
  15. 15.
    Viola, P., Jones, M.: Robust real-time object detection. International Journal of Computer Vision (2001)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  • Ignazio Gallo
    • 1
  • Marco Vanetti
    • 1
  • Simone Albertini
    • 1
  • Angelo Nodari
    • 1
  1. 1.Dipartimento di Scienze Teoriche e ApplicateUniversità dell’InsubriaVareseItaly

Personalised recommendations