A Real-Time Big Data Architecture for Glasses Detection Using Computer Vision Techniques

Automatic glasses detection is a hot topic withing the large-scale face images classification domain, which has impact on face recognition or soft biometrics for person identification. In many practical video surveillance applications, the faces acquired by cameras are low resolution. Therefore, this type of applications requires processing of a large number of relatively small-sized images. However, continuous stream of image and video data processing is a big data challenge. This need fits with the goals of Big Data streaming processing systems. In this paper, we propose a real-time Big Data architecture in order to collect, maintain and analyze massive volumes of images related with the problem of automatic glasses detection. This architecture can be used as an automatic image tagging related with glasses detection on face images.