While data-driven approaches could model complex sensor patterns, the required amount of labeled data increases exponentially as the number of contact points or the number of sensor subelements increase. We propose a novel deep learning framework that only needs single touch data to calibrate multiple contact points into a high resolution pressure map. The individual sub-local networks eliminate domain shift problems, while a message passing mechanism enables them to correctly learn correlation between neighboring sensor subelements.  Experimental results show that our framework can expand prior knowledge of single touch data to calibrate multi-touch sensor inputs into high resolution pressure maps.


Related publications

1. M Kim, H Choi, K-J Cho, S Jo, Single to Multi: Data-driven High Resolution Calibration Method for Piezoresistive Sensor Array, IEEE RA-L, 6(3): 4970-4977, July 2021  [LINK] [PDF] [VIDEO]