Off-screen Gesture Recognition Accuracy via Dynamic Condition Adjustment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing mobile terminals face low recognition rates for off-screen gestures due to misjudgment of similar gestures, leading to undesired operations or lack of response, as users can freely draw gestures on the touch screen, causing confusion and inefficiency.
Innovation Solution
A method that monitors the on/off state of off-screen gestures, matches them with a confusable gesture group, and adjusts recognition conditions to differentiate between intended and unintended gestures, ensuring only switched-on gestures trigger corresponding actions, thereby preventing misrecognition and improving recognition accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If the Off-screen gesture function allows users to freely draw gestures on the touch screen, then the ease of operation is improved, but the recognition accuracy deteriorates due to misjudgment of similar gestures
Solution Approach 1:
The patent applies local quality by differentiating gesture recognition based on specific characteristics of confusable gestures. Instead of using a single uniform recognition threshold, the system identifies gestures within confusable gesture groups and applies adjusted recognition conditions specifically to those gestures, allowing flexible input while maintaining accurate differentiation between similar gestures.
Solution Approach 2:
The patent implements dynamics by making the recognition condition adjustable based on the gesture type. The system dynamically changes the recognition condition for gestures belonging to confusable gesture groups compared to other gestures. This allows the system to adapt its strictness and sensitivity based on the specific gesture being recognized, balancing ease of operation with recognition accuracy.
2Device complexity
If the system uses a fixed recognition condition for all gestures, then the device complexity is reduced, but the recognition accuracy for similar gestures deteriorates
Solution Approach 1:
The patent applies segmentation by dividing gestures into different categories: confusable gesture groups and non-confusable gestures. The system segments the gesture recognition process into different handling paths based on whether a gesture belongs to a confusable group. This segmentation allows the system to maintain relatively simple processing for most gestures while applying enhanced differentiation only where needed, balancing complexity and accuracy.
3Productivity
If the system executes operations for all detected gestures, then the productivity is improved, but the loss of information increases due to executing undesired operations
Solution Approach 1:
The patent implements preliminary action by pre-identifying which gestures belong to confusable gesture groups before executing operations. The system预先 (in advance) determines the gesture category and applies appropriate recognition conditions beforehand. This preliminary classification prevents undesired operations from being executed by ensuring that only appropriately recognized gestures trigger corresponding functions, thereby avoiding loss of information and erroneous actions.
Data Source
AI summary
Embodiments provide a method, device, storage medium and mobile terminal, for recognizing an Off-screen gesture. The method includes: upon reception in a driver layer of an Off-screen gesture input by a user, determining whether each of preset Off-screen gestures is switched on based on an Off-screen gesture switch that switches on or off the preset Off-screen gesture in the driver layer; matching an switched-on Off-screen gesture with an Off-screen gesture in a preset confusable gesture group, and determining to change a recognition condition for the switched-on Off-screen gesture when the matching indicates there is only one Off-screen gesture in the on state in the confusable gesture group; and when determining to change the recognition condition, replacing, with a preset second recognition condition, a first recognition condition for the switched-on Off-screen gesture in the driver layer, and recognizing a gesture type of the input Off-screen gesture with the second recognition condition.