Application of 3D-wavelet statistics to video analysis

  • Authors:

  • M. Omidyeganeh;S. Ghaemmaghami;S. Shirmohammadi

  • Affiliations:

  • Electrical Engineering Department, Sharif University of Technology, Tehran, Iran and Advanced Information & Communication Technology Center (AICTC), Sharif University of Technology, Tehran, Iran a ...;Electrical Engineering Department, Sharif University of Technology, Tehran, Iran and Electronics Research Institute, Sharif University of Technology, Tehran, Iran;Distributed and Collaborative Virtual Environment Research Laboratory, University of Ottawa, Ottawa, Canada

  • Venue:
  • Multimedia Tools and Applications
  • Year:
  • 2013

Quantified Score

Hi-index 0.00

Visualization

Abstract

Video activity analysis is used in various video applications such as human action recognition, video retrieval, video archiving. In this paper, we propose to apply 3D wavelet transform statistics to natural video signals and employ the resulting statistical attributes for video modeling and analysis. From the 3D wavelet transform, we investigate the marginal and joint statistics as well as the Mutual Information (MI) estimates. We show that marginal histograms are approximated quite well by Generalized Gaussian Density (GGD) functions; and the MI between coefficients decreases when the activity level increases in videos. Joint statistics attributes are applied to scene activity grouping, leading to 87.3% accurate grouping of videos. Also, marginal and joint statistics features extracted from the video are used for human action classification employing Support Vector Machine (SVM) classifiers and 93.4% of the human activities are properly classified.