Learning Sparse Prototypes for Crowd Perception via Ensemble Coding Mechanisms

  • Yanhao Zhang
  • Shengping Zhang
  • Qingming Huang
  • Thomas Serre
Conference paper

DOI: 10.1007/978-3-319-11839-0_8

Part of the Lecture Notes in Computer Science book series (LNCS, volume 8749)
Cite this paper as:
Zhang Y., Zhang S., Huang Q., Serre T. (2014) Learning Sparse Prototypes for Crowd Perception via Ensemble Coding Mechanisms. In: Park H.S., Salah A.A., Lee Y.J., Morency LP., Sheikh Y., Cucchiara R. (eds) Human Behavior Understanding. HBU 2014. Lecture Notes in Computer Science, vol 8749. Springer, Cham

Abstract

Recent work in cognitive psychology suggests that crowd perception may be based on pre-attentive ensemble coding mechanisms consistent with feedforward hierarchical models of visual processing. Here, we extend a biological model of motion processing with a new dictionary learning method tailored for crowd perception. Our approach uses a sparse coding model to learn crowd prototypes. Ensemble coding mechanisms are implemented via structural and local coherence constraints. We evaluate the proposed method on multiple crowd perception problems from collective or abnormal crowd detection to tracking individuals in crowded scenes. Experimental results on crowd datasets demonstrate competitive results on par or better than state-of-the-art approaches.

Keywords

Sparse coding Crowd perception Biological vision 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  • Yanhao Zhang
    • 1
    • 2
  • Shengping Zhang
    • 1
  • Qingming Huang
    • 2
  • Thomas Serre
    • 1
  1. 1.Department of Cognitive Linguistic & Psychological Sciences, Institute for Brain SciencesBrown UniversityUSA
  2. 2.School of Computer ScienceHarbin Institute of TechnologyHarbinChina

Personalised recommendations