Gesture Keyboard Stroke Matching with Dwell Point Polyline Templates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Typical touch-screen based mobile computers with limited screen space face difficulties in fast and accurate data or text input using virtual keyboards, as users struggle to enter information efficiently without compromising accuracy.
Innovation Solution
A method and system that analyze gesture strokes on a virtual keyboard by identifying dwell points, generating a simplified polyline, and comparing it with predefined stroke templates to determine the corresponding word, facilitating efficient word input by matching the gesture stroke with the closest matching polyline.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If users input data via strokes on a limited screen space, then text input capability is enabled, but input speed and accuracy deteriorate
Solution Approach 1:
The patent segments the gesture recognition process into distinct phases: stroke detection, dwell point identification, polyline generation, and template matching. This segmentation allows each phase to be optimized independently, improving overall input efficiency while maintaining accuracy on limited screen spaces
Solution Approach 2:
The patent changes parameters by identifying dwell points (pauses in gesture movement) and generating simplified polylines from these points. This parameter transformation converts complex continuous gestures into discrete, comparable geometric representations, enabling faster matching with predefined word templates while preserving input accuracy
2Ease of operation
If users input data via strokes on a limited screen space, then text input capability is enabled, but input accuracy deteriorates
Solution Approach 1:
The patent applies preliminary action by pre-defining polyline templates for all possible words before user input. During gesture recognition, the system compares the user's simplified gesture polyline against these pre-prepared templates, enabling accurate word identification without requiring complex real-time analysis, thus maintaining high accuracy on limited screens
Solution Approach 2:
The patent uses copying by creating a simplified polyline representation (a geometric copy) of the user's gesture stroke. This copied representation captures the essential spatial characteristics of the gesture without all its noise and variations, enabling accurate comparison with template polylines for precise word identification
Data Source
AI summary
A method is provided for determining a word input by a gesture stroke on a virtual keyboard. The method includes receiving data representing the gesture stroke, analyzing the data to identify a set of dwell points in the gesture stroke, generating a simplified stroke defining a polyline having vertices corresponding with the dwell points of the identified set. The method further includes comparing the simplified stroke polyline with a plurality of predefined polylines each representing a respective word, to determine a closest matching polyline. The computing system outputs the word represented by the closest matching polyline.