ShoeSense: a new perspective on gestural interaction and wearable applications
Proceedings of the SIGCHI Conference on Human Factors in Computing Systems
| Hi-index | 0.00 |
We propose a novel hand-gesture recognition method based on mechanomyograms (MMGs). Skeletal muscles generate sounds specific to their activity. By recording and analyzing these sounds, MMGs provide means to evaluate the activity. Previous research revealed that specific motions produce specific sounds enabling human motion to be classified based on MMGs. In that research, microphones and accelerometers are often used to record muscle sounds. However, environmental conditions such as noise and human motion itself easily overwhelm such sensors. In this paper, we propose to use piezoelectric-based sensing of MMGs to improve robustness from environmental conditions. The preliminary evaluation shows this method is capable of classifying several hand gestures correctly with high accuracy under certain situations.