Recognition of Human Silhouette Based on Global Features

Recognition of Human Silhouette Based on Global Features

Milene Arantes (University of São Paulo, Brazil) and Adilson Gonzaga (University of São Paulo, Brazil)
Copyright: © 2012 |Pages: 10
DOI: 10.4018/978-1-4666-1574-8.ch021
OnDemand PDF Download:
List Price: $37.50
10% Discount:-$3.75


The aim of this paper is people recognition based on their gait. The authors propose a computer vision approach applied to video sequences extracting global features of human motion. From the skeleton, the authors extract the information about human joints. From the silhouette and the authors get the boundary features of the human body. The binary and gray-level-images contain different aspects about the human motion. This work proposes to recover the global information of the human body based on four segmented image models and applies a fusion model to improve classification. The authors consider frames as elements of distinct classes of video sequences and the sequences themselves as classes in a database. The classification rates obtained separately from four image sequences are then merged together by a fusion technique. The results were then compared with other techniques for gait recognition.
Chapter Preview

Traditional techniques for gait recognition are divided into two main groups: Methodologies based on Silhouette (Niyogi & Adelson, 1994; Hong, Lee, & Kim, 2007; Kale, Cuntor, Yegnanarayana, Rajagopalan, & Chellappa, 2003; Lee & Grimson, 2002) and methodologies based on Model (Bobick & Davis, 2001; BenAbdelkaer, Cutler, & Davis, 2002; Yang, Wu, & Peng, 2006; Liu & Zheng, 2007). In the approach based on silhouette, the shape and its derived attributes are the most important characteristics. Taking this into account, measurements are used to recognize a person in terms of the shape and/or movement.

The aim of the approach based on the Model is to interpret movement of the trunk and/or legs. In contrast to the approach based on silhouettes, generally the model-based methodology focuses on the dynamics of the movement, omitting the body shape. There are techniques to join the two models, thus providing a mixed methodology (Nizami et al., 2008; Wang, Tan, Ning, & Hu, 2003; Chellappa, RoyChowdhury, & Sundaresan, 2003).

Complete Chapter List

Search this Book: