Strategies of Dictionary Usages for Sparse Representations for Pedestrian Classification

  • Carlos Serra-Toro
  • Ángel Hernández-Górriz
  • V. Javier Traver
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10255)


Sparse representations and methodologies are currently receiving much interest due to their benefits in image processing and classification tasks. Despite the progress achieved over the last years, there are still many open issues, in particular for applications such as object detection which have been much less addressed from the sparse-representation point of view. This work explores several strategies for dictionary usages and study their relative computational and discriminative values in the binary problem of pedestrian classification. Specifically, we explore whether both class-specific dictionaries are really required, or just any of them can be successfully used and, in case that both dictionaries are required, which is the better way to compute the sparse representation from them. Results reveal that different strategies offer different computational-classification trade-offs, and while dual-dictionary strategies may offer slightly better performance than single-dictionary strategies, one of the most interesting findings is that just one class (even the negative, non-pedestrian class) suffices to train a dictionary to be able to discriminate pedestrian from background images.


Sparse coding Dictionary learning Pedestrian classification 


  1. 1.
    Castrodad, A., Sapiro, G.: Sparse modeling of human actions from motion imagery. Int. J. Comput. Vis. (IJCV) 100(1), 1–15 (2012)CrossRefGoogle Scholar
  2. 2.
    Dalal, N., Triggs, B.: Histograms of oriented gradients for human detection. In: CVPR (2005)Google Scholar
  3. 3.
    Deng, W., Hu, J., Guo, J.: In defense of sparsity based face recognition. In: CVPR (2013)Google Scholar
  4. 4.
    Dollár, P.: Piotr’s image and video matlab toolbox (PMT).
  5. 5.
    Dollár, P., Belongie, S., Perona, P.: The fastest pedestrian detector in the west. In: BMVC (2010)Google Scholar
  6. 6.
    Elad, M., Aharon, M.: Image denoising via learned dictionaries and sparse representation. In: CVPR (2006)Google Scholar
  7. 7.
    Fadili, M.-J., Starck, J.-L., Murtagh, F.: Inpainting and zooming using sparse. Comput. J. 52, 64–79 (2009)CrossRefGoogle Scholar
  8. 8.
    Fan, R.-E., et al.: LIBLINEAR: a library for large linear classification. J. Mach. Learn. Res. 9, 1871–1874 (2008)zbMATHGoogle Scholar
  9. 9.
    Liu, Y., Lasang, P., Siegel, M., Sun, Q.: Multi-sparse descriptor: a scale invariant feature for pedestrian detection. Neurocomputing 184, 55–65 (2016)CrossRefGoogle Scholar
  10. 10.
    Mairal, J., Bach, F., Ponce, J., Sapiro, G.: Online dictionary learning for sparse coding. In: ICML (2009)Google Scholar
  11. 11.
    Mairal, J., et al.: SPArse Modeling Software (SPAMS).
  12. 12.
    Ren, X., Ramanan, D.: Histograms of sparse codes for object detection. In: CVPR (2013)Google Scholar
  13. 13.
    Rigamonti, R., Brown, M., Lepetit, V.: Are sparse representations really relevant for image classification? In: CVPR (2011)Google Scholar
  14. 14.
    Shi, Q., Eriksson, A., van den Hengel, A., Shen, C.: Is face recognition really a compressive sensing problem? In: CVPR, (2011)Google Scholar
  15. 15.
    Tošić, I., Frossard, P.: Dictionary learning. IEEE Sig. Process. Mag. 28(2), 27–38 (2011)CrossRefGoogle Scholar
  16. 16.
    Wright, J., et al.: Robust face recognition via sparse representation. PAMI 31(2), 210–227 (2009)CrossRefGoogle Scholar
  17. 17.
    Wright, J., et al.: Sparse representation for computer vision and pattern recognition. Proc. IEEE 98(6), 1031–1044 (2010)CrossRefGoogle Scholar
  18. 18.
    Wright, J., Ganesh, A., Yang, A.Y., Zhou, Z., Ma, Y.: Sparsity and robustness in face recognition (2011). CoRR abs/1111.1014Google Scholar
  19. 19.
    Zheng, J., Jiang, Z., Chellappa, R.: Cross-view action recognition via transferable dictionary learning. IEEE Trans. Image Process. 25(6), 2542–2556 (2016)MathSciNetCrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  • Carlos Serra-Toro
    • 1
  • Ángel Hernández-Górriz
    • 1
  • V. Javier Traver
    • 1
  1. 1.Institute of New Imaging Technologies (INIT)Universitat Jaume ICastellón de la PlanaSpain

Personalised recommendations