A study on human observer classification and Ou-Isir database for gender and age estimation

Abstract

Identifying a person among a group of peoples is very difficult. Forthis, many techniques are introduced. Among which Gait identification is one of the best approaches. Because of its quality against unclear images, for a better data extraction silhouette image is used. This data like step size, foot length, walking hip moment etc., of a particularperson, is stored in the database to match it with obtained image. However, if we don’t have particular person data in the database, there is a probability of knowing the gender and age of that person. In this paper, initially, to identify the gender of a person from his silhouette image, the human observer’s classification of gender methodis used. This method includes a predictio n preference of some peoples are considered for identifying a person by his hair, chest, hip style, dressing, from a static image. After getting the gender, it is easy to estimate the age from the obtained image using OU- ISIR Database. In this work, we took two different groups of ages as children’s and adults.