Motion Input Confidence Threshold and User Confirmation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing motion-based input systems for computing devices often result in false positives due to incorrect matching of detected motions with stored input models, leading to unwanted actions, as they lack sufficient confidence thresholds and confirmation mechanisms.
Innovation Solution
A computing device with a processor configured to detect motion using a motion detection subsystem, compare the detected motion with a stored input model, and perform actions only if the confidence level exceeds a predetermined threshold or upon user confirmation, while also allowing for adjustment of the input model based on user feedback.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If motion-based input systems automatically perform actions based on detected motions, then the ease of operation is improved, but false positives occur leading to unwanted actions
Solution Approach 1:
The system implements feedback by detecting user confirmation or lack thereof when an action is triggered. If the user does not interact with the device within a threshold period after a motion is detected, the system infers negative confirmation and cancels the unwanted action. This feedback loop allows the system to learn from user responses and improve accuracy over time.
Solution Approach 2:
The system performs preliminary detection of motion patterns and compares them against stored input models before executing actions. By pre-establishing confidence thresholds and performing preliminary matching, the system can identify potential false positives before they result in unwanted actions, allowing for corrective measures.
2Reliability
If the system requires user confirmation before performing actions, then the reliability is improved, but the productivity decreases due to additional confirmation steps
Solution Approach 1:
The system applies partial confirmation requirements by only requiring user interaction when confidence levels are below the threshold or when false positives are detected. For high-confidence matches, the system performs actions automatically without requiring full confirmation steps, thus maintaining productivity while ensuring reliability when needed.
Solution Approach 2:
The confirmation requirement is dynamic rather than static. The system adjusts the level of confirmation needed based on the confidence level of the motion detection, the type of action being performed, and learned user preferences. This dynamic approach allows the system to require confirmation only when necessary, balancing reliability and productivity.
3Device complexity
If the system uses simple motion matching, then the device complexity is reduced, but the measurement precision decreases leading to incorrect matching
Solution Approach 1:
The system enhances measurement precision by adding temporal dimension to motion detection. Instead of only comparing spatial motion patterns, the system analyzes the timing, duration, and sequence of motions against stored input models. This temporal dimension allows for more accurate matching without significantly increasing device complexity, as it utilizes the existing motion detection subsystem more effectively.
Solution Approach 2:
The system changes parameters by introducing confidence thresholds and multiple comparison criteria rather than simple binary matching. By evaluating multiple parameters such as motion magnitude, direction, duration, and temporal patterns, the system achieves higher measurement precision while maintaining relatively simple device architecture through software-based analysis.
Data Source
Figure 1
Figure 2
Figure 3~4C
AI summary
A method implemented by a computing device having a memory and a motion sensor detection sensor is disclosed. The memory stores an input model associated with a gesture input. The method includes detecting a motion of the computing device; matching the detected motion with the input model; and determining a confidence level associated with the match. If the confidence level is above a pre-determined threshold, an action associated with the gesture input is performed automatically by the processor; and if the confidence level is below the pre-determined threshold, the action is performed by the processor responsive to receiving a positive confirmation of the gesture input.