Human Action Recognition by Extracting Features from Negative Space

  • Shah Atiqur Rahman
  • M. K. H. Leung
  • Siu-Yeung Cho
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6979)


A region based technique is proposed here to recognize human actions where features are extracted from the surrounding regions of a human silhouette termed as negative space. Negative space has the ability to describe poses as good as the positive spaces (i.e. silhouette based methods) with the advantage of describing poses by simple shapes. Moreover, it can be combined with silhouette based methods to make an improved system in terms of accuracy and computa-tional costs. Main contributions in this paper are two folded: proposed a method to isolate and discard long shadows from segmented binary images, and generalize the idea of negative space to work under viewpoint changes. The system consists of hierarchical processing of background segmentation, shadow elimination, speed calculation, region partitioning, shape based feature extraction and sequence matching by Dynamic Time Warping. The recognition accuracy of our system for Weizmann dataset is 100% and for KTH dataset is 95.49% which are comparable with state-of-the-art methods.


Human action recognition Negative space Silhouette Dynamic time warping complex activity fuzzy function 


  1. 1.
    Poppe, R.: A Survey on Vision-based Human Action Recognition. Image and Vision Computing 28(6), 976–990 (2010)CrossRefGoogle Scholar
  2. 2.
    Wang, L., Hu, W., Tan, T.: Recent Developments in Human Motion Analysis. Pattern Recognition 36(3), 585–601 (2003)CrossRefGoogle Scholar
  3. 3.
    Ikizler, N., Forsyth, D.A.: Searching for Complex Human Activities with No Visual Examples. IJCV 80(3), 337–357 (2008)CrossRefGoogle Scholar
  4. 4.
    Diaf, A., Ksantini, R., Boufama, B., Benlamri, R.: A novel human motion recognition method based on eigenspace. In: Campilho, A., Kamel, M. (eds.) ICIAR 2010. LNCS, vol. 6111, pp. 167–175. Springer, Heidelberg (2010)CrossRefGoogle Scholar
  5. 5.
    Wang, X., Ma, X., Grimson, W.E.: Unsupervised Activity Perception in Crowded and Complicated Scenes Using Hierarchical Bayesian Models. PAMI 31(3), 539–555 (2009)CrossRefGoogle Scholar
  6. 6.
    Lucena, M., de la Blanca, N.P., Fuertes, J.M., Marín-Jiménez, M.: Human action recognition using optical flow accumulated local histograms. In: Araujo, H., Mendonça, A.M., Pinho, A.J., Torres, M.I. (eds.) IbPRIA 2009. LNCS, vol. 5524, pp. 32–39. Springer, Heidelberg (2009)CrossRefGoogle Scholar
  7. 7.
    Rahman, S.A., Li, L., Leung, M.K.H.: Human Action Recognition by Negative Space Analysis. In: Cyberworlds (CW), pp. 354–359 (2010)Google Scholar
  8. 8.
    Li, L., Huang, W., Gu, I.Y.-H., Qi, T.: Statistical Modeling of Complex Backgrounds for Foreground Object Detection. Image Processing 13(11), 1459–1472 (2004)CrossRefGoogle Scholar
  9. 9.
    Prokop, R.J., Reeves, A.P.: A survey of moment-based techniques for unoccluded object representation and recognition. CVGIP 54(5), 438–460 (1992)Google Scholar
  10. 10.
    Gorelick, L., Blank, M., Shechtman, E., Irani, M., Basri, R.: Actions as Space-Time Shapes. PAMI 29(12), 2247–2253 (2007)CrossRefGoogle Scholar
  11. 11.
    Schuldt, C., Laptev, I., Caputo, B.: Recognizing Human Actions: A Local SVM Approach. In: ICPR, pp. 32–36 (2004)Google Scholar
  12. 12.
    Kellokumpu, V., Zhao, G., Pietikinen, M.: Dynamic Textures for Human Movement Recognition. In: Int. Conference on Image and Video Retrieval, pp. 470–476 (2010)Google Scholar
  13. 13.
    Ikizler, N., Duygulu, P.: Histogram of Oriented Rectangles: A New Pose Descriptor for Human Action Recognition. Image and Vision Computing 27(10), 1515–1526 (2009)CrossRefGoogle Scholar
  14. 14.
    Fathi, A., Mori, G.: Action Recognition by Learning Mid-level Motion Features. In: CVPR, pp. 1–8 (2008)Google Scholar
  15. 15.
    Lin, Z., Jiang, Z., Davis, L.S.: Recognizing Actions by Shape-Motion Prototype Trees. In: ICCV, pp. 444–451 (2009)Google Scholar
  16. 16.
    Schindler, K., van Gool, L.: Action Snippets: How Many Frames Does Human Action Recognition Require? In: CVPR, pp. 1–8 (2008)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Shah Atiqur Rahman
    • 1
  • M. K. H. Leung
    • 2
  • Siu-Yeung Cho
    • 1
  1. 1.School of Computer EngineeringNanyang Technological UniversitySingapore
  2. 2.FICTUniversiti Tunku Abdul Rahman (Kampar)Malaysia

Personalised recommendations