Multi-pass Touch Contact Tracking for Gesture Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional touch signal processing in portable devices with touch sensors and touch-screen displays struggles to accurately track and interpret large or long gesture motions, leading to gesture recognition failures and mis-tracking issues, especially when users perform fast gestures or input text on virtual keyboards.
Innovation Solution
The implementation of multi-pass touch contact tracking, which uses nearest-neighbor contact mapping in both forward and reverse directions to predict and validate touch contact components, reducing computational complexity and improving tracking accuracy while avoiding mis-tracking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional static matching is used for touch contact tracking, then small or short gestures can be tracked accurately, but large or long gestures with fast motion cannot be tracked correctly
Solution Approach 1:
The patent applies dynamics by transitioning from static matching to dynamic prediction-based tracking. The system predicts future touch contact positions based on historical motion data and updates tracking in real-time as gestures progress. This allows the tracking system to adapt to varying gesture speeds and lengths, maintaining accuracy for both small static gestures and large dynamic gestures across different motion ranges.
2Adaptability or versatility
If tracking processing is overly sensitive to gesture motion, then large gestures can be detected, but mis-tracking occurs during text input with virtual keyboards
Solution Approach 1:
The patent applies parameter changes by dynamically adjusting tracking sensitivity parameters based on gesture characteristics. The system monitors motion patterns and adapts prediction algorithms to distinguish between intentional gestures and unintentional contact variations during text input. This allows high sensitivity for detecting large gestures while maintaining stability during precise text input operations through context-aware parameter adjustment.
3Device complexity
If conventional tracking is used for fast gestures, then processing is simpler, but gesture recognition fails due to insufficient motion sensitivity
Solution Approach 1:
The patent applies preliminary action by pre-calculating prediction models and motion patterns before gestures occur. The system maintains historical touch contact data and pre-computes prediction parameters that can be quickly applied during fast gestures. This preparation allows the system to rapidly track fast movements without complex real-time calculations, achieving both simplicity and high recognition accuracy for rapid gestures.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
In embodiments of multi-pass touch contact tracking, touch input sensor data is recognized as a series of components of a contact on a touch-screen display. The components can be determined to correlate to the contact based on multi-pass nearest-neighbor contact mapping that includes forward nearest-neighbor contact mapping of the components and reverse nearest-neighbor contact mapping of the components. The components can then be associated to represent a tracking of the contact. Subsequent components of the contact can also be determined and associated with the previous components of the contact to further represent the tracking of the contact.