Touch Gesture Denoising via Data Point Resampling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Mobile computing devices face challenges in accurately determining user intent from touch gestures due to noise in touch input data, which affects the fitting of curves and subsequent animation velocities in graphical user interfaces.
Innovation Solution
The device adjusts time and position values of touch data points to evenly space them, allowing for curve fitting using methods like linear or nonlinear least-squares to determine user intent from touch gestures such as flicks, swipes, and pinches, thereby improving gesture recognition and animation velocity matching.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If curve fitting is performed directly on raw touch gesture data points, then the processing speed is faster, but the measurement precision of user intent is reduced due to noise in the data
Solution Approach 1:
The patent applies preliminary denoising and resampling actions to the touch gesture data before curve fitting. Specifically, it uses a Savitzky-Golay filter to remove noise from the raw touch data points, and then resamples the data to evenly space them in time. This preliminary processing improves the accuracy of subsequent curve fitting and user intent determination without significantly impacting processing speed.
2Measurement precision
If touch data points are evenly spaced in time through adjustment, then the curve fitting accuracy is improved, but the processing time increases
Solution Approach 1:
The patent changes the time parameter of touch data points by adjusting their timestamps to achieve even spacing. This parameter transformation allows the data to be uniformly distributed in time, which significantly improves curve fitting accuracy. The adjustment is performed efficiently by calculating new time values based on the total duration and number of points, minimizing the computational overhead.
3Reliability
If noise filtering is applied to touch input data, then the gesture recognition accuracy is improved, but the device complexity increases
Solution Approach 1:
The patent replaces complex mechanical or hardware-based noise filtering mechanisms with a software-based Savitzky-Golay filter algorithm. This polynomial-based filtering approach effectively removes noise from touch gesture data while maintaining the integrity of the gesture signal. The filter operates entirely in software, avoiding the need for additional hardware components and keeping the device complexity manageable.
Data Source
AI summary
In one embodiment, a method includes determining a touch gesture that includes two or more data points that each correspond to a particular location on a touch screen and a particular point in time or a particular location; adjusting, for each of one or more of the data points, a time or a position value representing its particular point in time or location with respect to other ones of the data points; and based on the adjusted values, determining a user intent associated with the touch gesture.