Adaptive Gesture Data Downsampling for Wearable Recognition

Solution Overview

Problem

Gesture recognition algorithms for wearable devices are computationally intensive due to the need for dynamic time warping, which is particularly challenging for devices with low computing power, and require user-specific training procedures.

Innovation Solution

A gesture recognition device that downsamples gesture data to a fixed number of data points using a smoothing window, allowing for more accurate and less computationally intensive comparisons by matching the downsampled data to a model gesture developed during training, using pattern matching techniques.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If dynamic time warping is used to compare gestures performed at different speeds, then gesture recognition accuracy is improved, but computational intensity increases

Engineering Contradiction:

Improvegesture recognition accuracyVSAvoidcomputational intensity

Core Design Contradiction:

Measurement precisionVSPower

Solution Approach 1:

The patent applies preliminary action by downsampling gesture data to a fixed number of data points before performing pattern matching. This preprocessing step reduces the computational burden of subsequent operations while preserving essential gesture characteristics, thereby resolving the contradiction between accuracy and computational intensity.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent changes the parameter of data point count from variable to fixed through downsampling. By transforming the gesture data to a standardized number of points, the system enables efficient pattern matching without requiring computationally intensive dynamic time warping, thus improving power efficiency while maintaining recognition accuracy.

Inventive Principle:

Principle #35Parameter changes

2Measurement precision

If gesture data is processed with high computational algorithms, then gesture recognition accuracy is improved, but device complexity increases

Engineering Contradiction:

Improvegesture recognition accuracyVSAvoidalgorithm complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent applies preliminary action by downsampling gesture data to a fixed number of data points before performing pattern matching. This preprocessing step reduces the computational burden of subsequent operations while preserving essential gesture characteristics, thereby resolving the contradiction between accuracy and computational intensity.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent changes the parameter of data point count from variable to fixed through downsampling. By transforming the gesture data to a standardized number of points, the system enables efficient pattern matching without requiring computationally intensive dynamic time warping, thus improving power efficiency while maintaining recognition accuracy.

Inventive Principle:

Principle #35Parameter changes

3Measurement precision

If user-specific training procedures are implemented, then gesture recognition accuracy is improved, but training time increases

Engineering Contradiction:

Improvegesture recognition accuracyVSAvoidtraining time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent applies preliminary action by downsampling gesture data to a fixed number of data points before performing pattern matching. This preprocessing step reduces the computational burden of subsequent operations while preserving essential gesture characteristics, thereby resolving the contradiction between accuracy and computational intensity.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent changes the parameter of data point count from variable to fixed through downsampling. By transforming the gesture data to a standardized number of points, the system enables efficient pattern matching without requiring computationally intensive dynamic time warping, thus improving power efficiency while maintaining recognition accuracy.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS10747327B2Technologies for adaptive downsampling for gesture recognition

Publication Date: 2020.08.18 INTEL CORP

AI summary

Technologies for gesture recognition using downsampling are disclosed. A gesture recognition device may capture gesture data from a gesture measurement device, and downsample the captured data to a predefined number of data points. The gesture recognition device may then perform gesture recognition on the downsampled gesture data to recognize a gesture, and then perform an action based on the recognized gesture. The number of data points to which to downsample may be determined by downsampling to several different numbers of data points and comparing the performance of a gesture recognition algorithm performed on the downsampled gesture data for each different number of data points.