3D Gesture Library Filtering for User-Defined Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing gesture recognition systems lack the ability to distinguish between interesting and non-interesting gestures in a three-dimensional sensory space, and they require indirect input devices like keyboards and mice for complex user interactions, limiting their applicability to simple commands.
Innovation Solution
A system that filters gestures by comparing user-defined reference characteristics with actual gestures in 3D space, allowing customization of gesture interpretation through user-defined parameters and demonstrations, and includes a gesture filter to identify relevant gestures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If gesture recognition systems use indirect input devices like keyboards and mice, then complex user interactions can be facilitated, but the system loses intuitiveness and requires additional setup equipment
Solution Approach 1:
The patent extracts and removes the dependency on indirect input devices (keyboards, mice) from the gesture recognition system. By filtering out non-gesture inputs and processing only direct gesture data from sensors, the system achieves complex interactions through intuitive gesture commands alone, eliminating the need for supplementary input devices
Solution Approach 2:
The gesture recognition system is designed to handle multiple types of interactions (navigation, selection, confirmation, cancellation) using a single gesture-based interface. The filterable gesture data structure allows the same sensor system to support both simple and complex interactions without requiring different input devices for different functions
2Reliability
If the gesture recognition system processes all detected gestures, then comprehensive gesture recognition is achieved, but computational complexity increases
Solution Approach 1:
The patent segments the gesture processing pipeline into distinct stages: gesture detection, filtering based on criteria (hand size, speed, position), and selective processing of filtered gestures. This segmentation allows the system to process only relevant gestures in detail while quickly discarding non-interesting ones, reducing overall computational complexity while maintaining recognition reliability
Solution Approach 2:
The system performs partial processing on all detected gestures by applying quick filter criteria to eliminate obviously non-relevant gestures without full analysis. Only gestures that pass the filter undergo complete recognition processing, allowing the system to handle comprehensive gesture input while minimizing computational overhead through selective deep processing
3Adaptability or versatility
If the system recognizes all gesture types in 3D space, then gesture recognition capability is enhanced, but difficulty in distinguishing relevant gestures increases
Solution Approach 1:
The patent applies different filtering criteria and processing approaches to different regions and types of gestures in 3D space. By defining specific filters for hand size, movement speed, and spatial position, the system tailors its detection sensitivity to local gesture characteristics, making it easier to distinguish relevant gestures from irrelevant ones while maintaining broad recognition capability
Solution Approach 2:
The system performs preliminary filtering of gesture data before full recognition processing. By pre-evaluating gestures against established criteria (comparing actual gestures to reference gestures, checking hand velocity and position thresholds), the system prepares and organizes gesture data in advance, reducing the difficulty of identifying relevant gestures during main processing
Data Source
AI summary
The technology disclosed relates to the creation of a gesture library for subsequent use in filtering gestures. In particular, the methods disclosed here generate and store gestures and their characteristic values to create a set of user-defined reference gestures that can be compared against characteristics of actual gestures performed in a 3D sensory space. Based on these comparisons, a set of gestures of interest may be filtered from all the gestures performed in the 3D sensory space. The technology disclosed also relates to customizing gesture interpretation for a particular user, and to setting parameters for recognizing gestures by prompting the user to select values for characteristics of the gestures. In one implementation, the technology disclosed includes performing characteristic focused demonstrations of boundaries of the gesture. It further includes testing the interpretation of gestures by prompting the user to perform complete gesture demonstrations and receiving user evaluation regarding the interpretation.


