Using robust audio and video processing technologies to alleviate the elderly cognitive decline

Abstract

We are recently witnessing a growing interest for pervasive context-aware products and services for elderly users. This is largely due to falling fertility and rising longevity phenomena, as well as due to the proliferation of the aging population all over the world. In this paper we present a number of leading edge audio and video processing technologies, which can be exploited to build robust ambient assisted living applications for elderly groups. In particular, we discuss application requirements aiming at alleviating the cognitive decline of elderly users and present audio and video processing components that can essentially fulfill these requirements. We emphasize on technologies such as automatic speech recognition, speaker identification, face detection, person tracking, face identification, and demonstrate how mature versions of these technologies can be appropriately customized to give a significant boost to AAL applications for senior citizens. The challenges, solutions and ideas within this paper are part of the EU project HERMES, which aims at providing an integrated approach to cognitive care, based on assistive technology that reduces age-related decline of cognitive capabilities. Copyright 2008 ACM.