Gesture Recognition Scaling by User Physical Dimensions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional gesture recognition systems using mechanical devices like accelerometers are limited in accuracy as they prioritize acceleration over distance, leading to reduced interpretation of body movement and inaccuracy in gesture input, especially when gestures other than hand movements are involved.
Innovation Solution
An apparatus and method that obtain and utilize at least one physical dimension of a user, such as height or arm span, to determine gestures independently of the user's location relative to the system, using sensors and a learning algorithm to calibrate and scale gestures for improved accuracy and fairness across different users.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional gesture recognition systems use mechanical devices like accelerometers to determine body motion, then the system can detect movement and acceleration, but the accuracy of gesture interpretation is reduced because the system prioritizes acceleration over distance traveled
Solution Approach 1:
The patent changes the parameters used for gesture recognition from acceleration-based to distance-based measurements. By integrating accelerometer data over time to calculate distance traveled and using this distance information as the primary gesture parameter, the system achieves more accurate gesture interpretation that accounts for the full range of motion rather than just peak acceleration events.
2Measurement precision
If the system relies on acceleration data from accelerometers, then movement detection is possible, but the distance traveled by the gesture is considered less important leading to reduced accuracy
Solution Approach 1:
The system performs preliminary integration of acceleration data to calculate velocity and position before gesture recognition occurs. By pre-processing the accelerometer data to extract distance traveled information, the system makes this information available for accurate gesture interpretation without losing the relationship between acceleration and displacement.
Solution Approach 2:
The patent introduces distance traveled as an intermediary parameter that mediates between raw acceleration data and gesture interpretation. This intermediate calculation of displacement provides a more meaningful representation of body movement that bridges the gap between sensor output and gesture meaning.
3Measurement precision
If gesture recognition does not account for individual physical dimensions, then the system is simpler to implement, but accuracy is reduced for users with different body sizes
Solution Approach 1:
The patent applies local quality by customizing gesture recognition parameters for each individual user based on their specific physical dimensions. The system calibrates distance thresholds and gesture parameters to match each user's body size, arm length, and movement characteristics, ensuring accurate recognition for users of all sizes rather than using a one-size-fits-all approach.
Solution Approach 2:
The system performs preliminary calibration to obtain physical dimension information from each user before gesture recognition begins. This advance collection of user-specific measurements allows the system to pre-adjust gesture parameters and thresholds, eliminating the need for complex real-time adjustments during gesture input.
Data Source
AI summary
An apparatus for gesture recognition, according to aspects of the disclosure contained herein, include a processing system configured to obtain at least one physical dimension of a user and determine a gesture of the user based on the at least one physical dimension independent of a location of the user relative to the apparatus. A method for gesture recognition is also disclosed.