Exploiting scene and body contexts in controlling continuous vision body cameras

Ever-increasing performance at decreasing price has fueled camera deployments in a wide variety of real-world applications—making the case stronger for battery-powered, continuous-vision camera systems. However, given the state-of-the-art battery technology and embedded systems, most battery-powered mobile devices still do not support continuous vision. In order to reduce energy and storage requirements, there have been proposals to offload energy-demanding computations to the cloud Naderiparizi et al. (2016) [1], to discard uninteresting video frames Naderiparizi et al. (2017), and to use additional sensors to detect and predict when to turn on the camera Bahl et al. (2012) [2]. However, these proposals either require a fat communication bandwidth or have to sacrifice capturing of important events.