Head and shoulders pose estimation using a body-mounted camera
Hirotake Yamazoe · 2017
In this paper, we propose a wearable camera-based pose estimation of a user's head and shoulders. The proposed system is mounted on a user's chest and the camera looks upward to observe a user's head and shoulders region. The proposed method is based on histograms of orientated gradient (HoG) features and support vector regression (SVR). This paper shows the preliminary experimental results that demonstrate the effectiveness of the proposed method.