Low-Power and Cost-Effective Context Information Estimation for Wearables.

Context awareness helps Artificial Intelligence (AI) better communicate with humans. It has a wide application range from entertainment and marketing to education and healthcare. In this paper we propose our method for a low-power, low-cost, and versatile context-aware system based on chest-worn Inertial Measurement Units (IMUs). The ability to recognize subjective feelings enables the proposed system to perform affective computing next to activity recognition, posture recognition, swallow detection, and voice activity detection. We set up a number of experiments and collected relevant data on which we are currently performing our methods.