Gesture Recognition Using Compressed Feature Matrices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current gesture recognition systems face challenges in accurately and efficiently differentiating between various forms of human body movements for communication and entertainment, particularly in real-time applications, due to the complexity and subjectivity of defining gestures and the diversity of human mannerisms.
Innovation Solution
The development of a machine learning-based system that uses self-referenced gesture data compressed by principal joint variable analysis and principal component analysis to identify and classify gestures, employing a classifier and recognizer to compare incoming data against a database of learned movements, allowing for efficient recognition and prediction of gestures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional shape descriptors and various processes are used for gesture recognition, then gesture differentiation capability is improved, but system complexity and computational overhead increase
Solution Approach 1:
The patent extracts only the most relevant features from gesture data by identifying and removing redundant features. The system determines a subset of features that are sufficient for gesture recognition, eliminating unnecessary computational processing of redundant data while maintaining recognition accuracy.
Solution Approach 2:
The patent creates simplified representations of gesture data by generating synthetic gesture data that captures essential characteristics without requiring complex original data processing. This allows the system to work with compressed feature sets that replicate the information needed for accurate gesture differentiation.
2Measurement precision
If comprehensive gesture data is collected and processed, then gesture recognition accuracy is improved, but processing time increases
Solution Approach 1:
The patent extracts only the essential features needed for gesture recognition by determining a sufficient subset of features from the complete gesture data set. This extraction process eliminates redundant computational steps while preserving the information necessary for accurate gesture identification.
Solution Approach 2:
The patent performs preliminary processing by pre-determining which features are sufficient for gesture recognition before actual gesture classification occurs. This advance preparation of feature subsets reduces the computational burden during real-time gesture processing.
3Measurement precision
If detailed gesture data is analyzed, then gesture classification accuracy is improved, but data processing complexity increases
Solution Approach 1:
The patent identifies and extracts only the relevant features from detailed gesture data that are necessary for accurate classification. By determining a sufficient subset of features, the system reduces data processing complexity while maintaining the ability to accurately differentiate between various gesture types.
Data Source
AI summary
Systems and method described herein present techniques for identifying a gesture using gesture data compressed by principal joint variable analysis. A classifier of a gesture recognition system may receive a frame comprising a set of gesture data points identifying locations of body parts of a subject. The classifier may determining that a subset of the set of gesture data points is sufficient to recognize a first gesture. The subset may be stored into a database in reference to the first gesture. A recognizer may receive a new frame of new gesture data points identifying locations of body parts of a new subject. The recognizer may recognize that the gesture of the new subject corresponds to the first gesture responsive to comparing at least one new gesture data point from the new frame to at least one gesture data point of the subset.