Automatic gait recognition based on probabilistic approach

Imran Fareed Nizami, Sungjun Hong, Heesung Lee, Byungyun Lee, Euntai Kim

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

A simple probabilistic method for online video based human identification is introduced in this article. The proposed method is based on a modified version of Motion Silhouette images (MSI) and recursive probability accumulation. The modified version of MSI is named the Moving Motion Silhouette Image (MMSI). Identification probability is accumulated recursively in a Bayesian framework to draw a single conclusion from the whole gait sequence. The probability is named the accumulated posterior probability (APP) and denotes the probability based on all the information available up to now. The proposed method is tested on the well-known publicly available NLPR and SOTON gait databases. The experimental results demonstrate the effectiveness of the proposed algorithm and indicate the fact that using MMSI and APP for information fusion yields higher recognition rates as compared to previous gait recognition systems.

Original languageEnglish
Pages (from-to)400-408
Number of pages9
JournalInternational Journal of Imaging Systems and Technology
Volume20
Issue number4
DOIs
Publication statusPublished - 2010 Dec 1

All Science Journal Classification (ASJC) codes

  • Electronic, Optical and Magnetic Materials
  • Software
  • Computer Vision and Pattern Recognition
  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'Automatic gait recognition based on probabilistic approach'. Together they form a unique fingerprint.

  • Cite this