Authors: Minsu Jang; Jaehong Kim; Yong-Ho Seo; Hyun-Seung Yang
Addresses: Department of Computer Science, Korea Advanced Institute of Science and Technology (KAIST), Daejeon-si, South Korea; Electronics and Telecommunications Research Institute (ETRI), Daejeon-si, South Korea ' Electronics and Telecommunications Research Institute (ETRI), Daejeon-si, South Korea ' Department of Intelligent Robot Engineering, Mokwon University, Daejeon-si, South Korea ' Department of Computer Science, Korea Advanced Institute of Science and Technology (KAIST), Daejeon-si, South Korea
Abstract: We present in this paper an adaptive gesture classifier for mobile devices, along with an efficient method to automatically detect endpoints of gestures. A classification model based on 1-NN with DTW-based k-means clustering is augmented by a metacognitive framework that measures the quality of the learned model and continuously updates it to improve the performance. We evaluated the model with an accelerometer signal database of 26 English alphabets. The results showed that the adaptive framework improved the recall and precision rates by 4.9% and 5.6%, respectively. Our endpoint detection method, based on energy variance and low-pass filtering, successfully detected 98.5% of gestures with an average detection delay of 176 ms.
Keywords: accelerometers; gesture recognition; adaptation; metacognition; adaptive gesture tracking; acceleration sensors; mobile devices; DTW; discrete wavelet transform; k-means clustering.
International Journal of Wireless and Mobile Computing, 2015 Vol.8 No.2, pp.183 - 193
Received: 19 Nov 2013
Accepted: 20 Aug 2014
Published online: 28 Mar 2015 *