Precision Gesture-based Inputs for Smartphone with Two Consecutive Simple Gestures

摘要

In recent years, smartphones and other mobile computers have become even smaller. Simultaneously, the number of buttons for user input has decreased, and their size has shrunk, which lowers the operability of user input. In this situation, gesture-based input is expected to be the new input method. gesture-based inputs have advantage in that it can work irrespective of device size Furthermore, there is no need to even look at the screen. Accelerometers equipped in smartphones are the primary tool for recognizing gesture-based inputs. However, such techniques are prone to misrecognitions. One factor contributing to this problem is the high number of gesture candidates. In response, we propose an input recognition method that uses a combination of two far simpler gestures types: trigger gesture (TG) and main gesture (MG). TG determine the start or content of the operation. MG provide the operational detail. In this paper, we test this concept on a music player app installed on a smartphone, which can be used while walking. The proposed method is found to improve recognition accuracy by 50%, assuming five types of input requests.