Hand Gesture Detection Using Discrete Values and SSVM
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current hand gesture recognition technologies are immature, lacking accuracy and complexity, requiring high computation resources and relying on full hand skeleton articulation and complex machine learning algorithms, making them costly and unsuitable for widespread adoption.
Innovation Solution
The system employs a discrete values architecture to represent hand gestures, using structured support vector machine functions and finite state machines to estimate runtime hand datasets, reducing computation load by quantizing continuous values into discrete hand values and optimizing hand gesture detection through dynamic programming.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If full hand skeleton articulation and complex machine learning algorithms are used for hand gesture detection, then detection accuracy may be improved, but computational resources and system complexity increase significantly
Solution Approach 1:
The patent segments the hand gesture recognition problem into distinct phases: hand detection, landmark identification, and gesture classification. Each phase processes only relevant features rather than full skeleton data, reducing computational load while maintaining accuracy. The system divides hand features into key landmarks (tip, middle, base of fingers) and processes them separately through specialized algorithms.
Solution Approach 2:
The patent extracts only the essential features needed for gesture recognition from the full hand skeleton. Instead of processing all 21 hand joints, the system identifies and extracts 10 key landmarks that capture the essential geometry of hand poses. This extraction approach maintains detection accuracy while significantly reducing the data volume requiring complex processing.
2Measurement precision
If full hand skeleton articulation and complex machine learning algorithms are used for hand gesture detection, then detection accuracy may be improved, but computational resources increase
Solution Approach 1:
The patent applies partial action by using simplified geometric models and lightweight algorithms for the majority of processing cases. The system employs efficient distance calculations and geometric transformations for common gesture patterns, reserving complex machine learning operations only for ambiguous or difficult cases. This approach reduces average computational resource consumption while maintaining high accuracy.
Solution Approach 2:
The patent replaces heavy machine learning models with geometric and algebraic methods for feature extraction and gesture classification. The system uses coordinate transformations, distance metrics, and geometric relationships to interpret hand poses, substituting computationally intensive neural networks with more efficient mathematical operations that consume fewer computational resources.
3Reliability
If current hand gesture recognition technologies are used, then some basic gestures may be detected, but accuracy is insufficient and complexity is high
Solution Approach 1:
The patent performs preliminary classification of hand gestures into categories (pinch, grasp, open hand, etc.) before detailed analysis. The system first identifies the gesture type using simplified criteria, then applies specialized processing only for that category. This preliminary action reduces the complexity of subsequent processing by narrowing down the problem space early in the recognition pipeline.
Solution Approach 2:
The patent transforms hand landmark coordinates into normalized geometric parameters such as finger lengths, angles, and relative positions. By changing the representation from raw coordinates to normalized parameters, the system achieves scale and rotation invariance, improving recognition accuracy while simplifying the comparison process through standardized feature vectors.
Data Source
AI summary
Computer implemented method for detecting a hand gesture of a user, comprising:(a) Receiving sequential logic models each representing a hand gesture. The sequential logic model maps pre-defined hand poses and motions each represented by a hand features record defined by discrete hand values each indicating a state of respective hand feature.(b) Receiving a runtime sequence of runtime hand datasets each defined by discrete hand values scores indicating current state hand features of a user's moving hand which are inferred by analyzing timed images depicting the moving hand.(c) Submitting the runtime hand datasets and the pre-defined hand features records in SSVM functions to generate estimation terms for the runtime hand datasets with respect to the hand features records.(d) Estimating which of the hand gestures best matches the runtime sequence depicted in the timed images by optimizing score functions using the estimation terms for the runtime hand datasets.