Refined Segmentation of Images for Human Pose Analysis

Image-based gait analysis as a means of biometric identification has attracted much research attention. Most of the existing methods focus on human posture identification and tracking. There have been few investigations on the relationship between the carried load by a person and the change of gait characteristics. Nevertheless, this relationship can be very useful in a number of applications, such as studying the postural effects of load on children and adolescence. In this paper, we investigate how to estimate carried weight from a sequence of images of a person walking normally. Observing that human tends to minimize energy expenditure during walking, we compute several angles of body leaning and determine the relationship among the carried weight, the leaning angles, and the location of the center of gravity. This method has been verified successfully by experiments.