Human age classification based on gait parameters using a Gait Energy Image projection model

With the increasing significance of age classification in present days, researchers are working on different methods to classify a persons' age. Facial based and Gait based are the major trail methods for age classification. Actually, the facial based approach is not so accurate if the person is far from the camera. Whereas, gait is a preferable solution because it is quick to respond to age parameters. In this paper, Gait energy image Projection model (GPM) is the proposed method for age classification, which combines both spatiotemporal Gait energy image Longitudinal projection (GLP) and Gait energy image Transverse Projection (GTP). The proposed method mainly focuses on four parameters namely head movement, body size, arm movement and Stride length. Regarding classification of age, OU-ISIR dataset is considered and the SVM is selected as the classifier. Moreover, obtained experimental results are compared with the existing ones like FED, GEI and SM. Further Descriptors are fused to check whether they give better results or not.