Gesture Recognition via Segmented Trajectory Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing gesture recognition methods require significant computing power, making them unsuitable for resource-constrained devices such as mobile devices in vehicle technology, where reliable and efficient gesture recognition is needed.

Innovation Solution

A method that captures image data from at least two individual images, segments the images to determine individual objects, calculates reference points, and tracks trajectories to identify gestures, using techniques like segmentation, trajectory analysis, and distance profiling to generate output signals without requiring extensive computational resources.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

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

Engineering Contradiction:
Improvegesture recognition reliabilityVSAvoidcomputing power consumption
Core Design Contradiction:
ReliabilityVSUse of energy by moving object

Solution Approach 1:

The patent segments the gesture recognition process into distinct phases: image acquisition, preprocessing, feature extraction, and gesture determination. By dividing the complex recognition task into smaller, manageable segments, the system achieves reliable gesture recognition while reducing overall computational burden on resource-constrained devices.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary actions by pre-processing images and extracting features before final gesture determination. Reference points are calculated in advance from segmented objects, and trajectories are pre-computed from sequences of reference points. This preliminary processing reduces the computational load during real-time gesture recognition.

Inventive Principle:
Principle #10Preliminary action

2Measurement precision

If complex gesture analysis is performed, then gesture recognition accuracy is improved, but processing speed decreases

Engineering Contradiction:
Improvegesture detection accuracyVSAvoidprocessing speed
Core Design Contradiction:
Measurement precisionVSSpeed

Solution Approach 1:

The patent extracts only the essential features needed for gesture recognition: reference points from segmented objects and trajectories from reference point sequences. By taking out and focusing on these critical elements rather than analyzing all image data, the system maintains high detection accuracy while achieving fast processing speeds suitable for real-time applications.

Inventive Principle:
Principle #2Taking out (Extraction)

3Reliability

If comprehensive image processing is used, then gesture recognition reliability is improved, but device complexity increases

Engineering Contradiction:
Improvegesture recognition reliabilityVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent applies local quality by performing segmentation and feature extraction only on relevant regions of the image data. Instead of processing entire images uniformly, the system focuses computational resources on detecting and analyzing specific objects and their trajectories, thereby achieving reliable gesture recognition with reduced system complexity.

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP3642696B1Method and device for detecting a user input on the basis of a gesture
Publication Date: 2023.09.06 VOLKSWAGEN AG
  • EP3642696B1 patent drawingFigure 1~2A
  • EP3642696B1 patent drawingFigure 2B~2C
  • EP3642696B1 patent drawingFigure 3A~4B

AI summary

The invention relates to a method for detecting a user input on the basis of a gesture, in which method image data of at least two individual images are acquired, recoding times being allocated to the individual images. Each of the acquired individual images is segmented, an individual image object (31) is identified in each of the individual images and a reference point (33, 39) is determined on the basis of the individual image object (31). A trajectory (40) is determined on the basis of the reference points (33, 39) in the individual images and a gesture is determined on the basis of the trajectory (40). 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, comprising an acquisition unit (4) for acquiring image data, a segmentation unit (5) for performing segmentation, a trajectory computing unit (10) for determining the trajectory (40), an allocation unit (7) for determining a gesture and an output unit (8).