Gesture Recognition via Measured Distance Profile

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing gesture recognition methods require significant computing power, making them unsuitable for mobile devices like those in vehicle technology, where reliable and efficient gesture detection is needed with limited resources.

Innovation Solution

A method that uses image data to determine a measured distance profile from an object's outline, allowing for quick and resource-efficient gesture recognition, even with low-resolution images, by segmenting the image, determining a reference point, and generating a distance profile based on object distances, which can be used to identify gestures with minimal computational effort.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional gesture recognition methods are used, then gesture recognition reliability is improved, but computing power requirements increase significantly

Engineering Contradiction:
Improvegesture recognition reliabilityVSAvoidcomputing power requirements
Core Design Contradiction:
ReliabilityVSPower

Solution Approach 1:

The patent segments the gesture recognition process into distinct stages: image acquisition, preprocessing, feature extraction (distance profile generation), and gesture classification. By dividing the complex recognition task into smaller, manageable segments, the system achieves reliable gesture recognition while reducing the computational burden on mobile devices. The distance profile segmentation specifically extracts only the essential geometric features needed for recognition.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent extracts and focuses on a specific, simplified feature representation called the distance profile, which captures the essential geometric characteristics of gestures. Instead of processing entire images or using complex deep learning models, the system extracts only the necessary distance measurements from reference points, significantly reducing computing power requirements while maintaining recognition reliability.

Inventive Principle:
Principle #2Taking out (Extraction)

2Measurement precision

If high-resolution image data is used for gesture recognition, then recognition precision is improved, but processing time and computational effort increase

Engineering Contradiction:
Improvegesture recognition precisionVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent extracts only the essential geometric information needed for gesture recognition by generating distance profiles from reference points. This extraction approach captures the critical spatial relationships without processing the entire high-resolution image data, thereby maintaining recognition precision while significantly reducing processing time and computational effort.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent uses low-resolution or simplified image data that can be quickly processed and discarded after feature extraction. Instead of relying on computationally expensive high-resolution images, the system uses simplified representations (distance profiles) that require minimal processing time while still providing sufficient precision for accurate gesture recognition.

Inventive Principle:
Principle #27Cheap short-living objects (Disposable)

Data Source

PatentEP3642697B1Method and device for detecting a user input on the basis of a gesture
Publication Date: 2023.08.23 VOLKSWAGEN AG

AI summary

In the method according to the invention for detecting a user input on the basis of a gesture, image data are acquired and segmentation is performed on the basis of the acquired image data, determining an object (31) with an outline (32). A reference point (33) is determined on the basis of the object (31), a distance between each of a plurality of points located on the outline (32) of the object (31) and the reference point (33) is determined and a measured distance profile (37) is generated on the basis of these distances. A gesture is determined on the basis of the measured distance profile (37) and an output signal is generated and output on the basis of the gesture determined. The invention further relates to a device for detecting a user input on the basis of a gesture. The device comprises an acquisition unit (4) for acquiring image data and a segmentation unit (5) for performing segmentation on the basis of the image data acquired, thereby determining an object (31) having an outline (32). The device further comprises a profile computing unit (6) for determining a reference point (33) on the basis of the object (31), wherein a distance between each of a plurality of points located on the outline (32) of the object (31) and the reference point (33) can be determined and a measured distance profile (37) can be generated on the basis of these distances. The device comprises an allocation unit (7) by means of which a gesture can be determined on the basis of the measured distance profile (37) and an output unit (8) by means of which an output signal can be generated and output on the basis of the gesture determined.