TOF Gesture Recognition Using MAD Calculation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current hand gesture recognition systems for electronic devices require intensive processing and memory, making them unsuitable for smaller devices like cell phones and tablets, and are inefficient in distinguishing between common gestures such as taps and wipes.

Innovation Solution

An electronic device equipped with a laser source and detectors uses time-of-flight measurements and Mean Absolute Deviation (MAD) values to differentiate between hand gestures like taps and wipes, employing a controller that calculates MAD values and identifies gestures with reduced processing power, utilizing a single integrated circuit with a Vertical-Cavity Surface-Emitting Laser (VCSEL) and Single Photon Avalanche Diode (SPAD) detectors.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If complicated algorithms including time-of-flight and machine learning are used to distinguish hand gestures, then gesture recognition accuracy is improved, but processing power and memory requirements increase excessively

Engineering Contradiction:

Improvegesture recognition accuracyVSAvoidprocessing power and memory requirements

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts only the essential feature needed for gesture differentiation - the Mean Absolute Deviation (MAD) of distance values - from the complex algorithmic processing. By focusing solely on MAD calculation rather than full machine learning algorithms, the system achieves adequate gesture recognition with minimal processing requirements, effectively extracting only what is necessary from the complex processing approach.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

Instead of using complex algorithms to directly classify gestures, the patent inverts the approach by first calculating simple statistical properties (MAD) of the distance data and then using these simplified metrics for gesture identification. This inversion from direct complex classification to indirect simplified metric-based classification resolves the contradiction between accuracy and processing complexity.

Inventive Principle:

Principle #13The other way round (Inversion)

2Device complexity

If simple infrared sensors are used to detect hand gestures, then device complexity is reduced, but gesture discrimination capability between taps and wipes deteriorates

Engineering Contradiction:

Improvesensor system simplicityVSAvoidgesture discrimination capability

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The patent changes the parameter used for gesture discrimination from simple brightness detection to the statistical parameter MAD (Mean Absolute Deviation) of time-of-flight distance values. This parameter change enables simple sensors to achieve better gesture discrimination by analyzing the variation pattern of distance measurements rather than relying on complex sensor arrays or algorithms.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent replaces the mechanical/approach of using complex sensor systems or algorithms with a computational substitution - using time-of-flight measurements combined with MAD calculation. This substitution achieves enhanced gesture discrimination capability while maintaining sensor system simplicity by leveraging computational processing of basic distance measurements rather than mechanical sensor complexity.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This solution enables efficient hand gesture recognition with lower processing power and memory usage, allowing for precise detection of gestures up to 50 cm away, comparable to more complex systems, while being suitable for compact devices like mobile wireless communications devices and consumer electronics.

Implementation Method 1

A controller is coupled to the laser source and at least one laser detector and configured to determine a plurality of distance values to the user's hand based upon a time-of-flight of the laser radiation

Methodology Applied

Scientific EffectTime-of-flight: Time of Flight

Implementation Method 2

The laser source may comprise a vertical-cavity surface-emitting laser (VCSEL)

Methodology Applied

Scientific EffectLaser: Laser

Implementation Method 3

The at least one laser detector may comprise a single photon avalanche diode (SPAD) detector configured to receive reflected laser radiation

Methodology Applied

Scientific EffectPhotoelectric Effect: Photoelectric Effect

Data Source

PatentUS10303254B2Device and method for identifying tap or wipe hand gestures using time-of-flight sensing

Publication Date: 2019.05.28 STMICROELECTRONICS FRANCE

AI summary

An electronic device includes a laser source configured to direct laser radiation toward a user's hand. A laser detector is configured to receive reflected laser radiation from the user's hand. A controller is coupled to the laser source and laser detector and configured to determine a plurality of distance values to the user's hand based upon a time-of-flight of the laser radiation, calculate a mean absolute deviation (MAD) value based upon the plurality of distance values, and identify whether the user's hand is moving in a first or second gesture based upon the MAD value.