Web image mining for facial age estimation

Abstract

Face image based age estimation is an approach to classify face images into one of several pre-defined age-groups. It is challenging because the aging variation is specific to a given individual and is determined by not only the person's gene, but also by many external factors, such as exposure, weather conditions (e.g. ambient humidity), health, gender, living style and living location. Age categorization is a multiclass problem. Lack of a good public aging dataset makes it difficult to estimate age from face image. Due to large multiclass nature of human ages, it is difficult to collect large aging database. In this paper, we proposed a novel approach to collect high quality aging face images. Application Programming Interface (API) services provided by Microsoft Search Engine Bing are adopted for this purpose. Automatic collection of age-specified query images was implemented in C# programming interface with Microsoft BING API services. The experimental results of age estimation has verified that our approach is a good way to collect aging images in huge volume since internet hosts a large number of images with age labeling. It serves as an alternative cheap platform to gather face images for age classification.