View-invariant action recognition using cross ratios across frames

  • Authors:

  • Yeyin Zhang;Kaiqi Huang;Yongzhen Huang;Tieniu Tan

  • Affiliations:

  • National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences;National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences;National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences;National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences

  • Venue:
  • ICIP'09 Proceedings of the 16th IEEE international conference on Image processing
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

We present a new method of computing invariants in videos captured from different views to achieve view-invariant action recognition. To avoid the constraints of collinearity or coplanarity of image points for constructing invariants, we consider several neighboring frames to compute cross ratios, namely cross ratios across frames (CRAF), as our invariant representation of action. For every five points sampled with different intervals from the trajectories of action, we construct a pair of cross ratios (CRs). Afterwards, we transform the CRs to histograms as the feature vectors for classification. Experimental results demonstrate that the proposed method outperforms the state-of-the-art methods in effectiveness and stability.