A recognition method for combined activities with accelerometers
Kazuya Murao, Tsutomu Terada · 2014
Many activity recognition systems using accelerometers have been proposed. Activities that have been recognized are "single" activities which can be expressed with one verb, such as sitting, walking, holding a mobile phone, and throwing a ball. In actual, however, "combined" activities including more than two kinds of state and movement are often taken place. Focusing on hand gestures, they are performed not only while standing, but also while walking and sitting. Though the simplest way to recognize such combined activities is to construct the recognition models for all the possible combinations of the activities, the number of combinations becomes immense. In this paper, we propose a recognition method for combined activities by learning single activities only. Evaluation results confirmed that our proposed method achieved 0.84 of recall and 0.85 of precision, which is comparable to the method that had learned all the combined activities.