Feature extraction methods for human gait recognition – A survey

Abstract

Human gait recognition opens a wide variety of challenging problems for research community. Feature extraction has a significant role in designing human gait recognition systems. Numerous features have been defined based on gait video frames. Spatial as well as temporal descriptors have equal importance within gait features. In this paper, we present a survey of prominent feature extraction methods incorporated in human gait recognition systems and their respective recognition accuracies are reported. Also, a description of popular gait databases is presented.