Vehicle Target Recognition Using Gaze and Hand Vector Intersection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for target recognition in vehicles require expensive calibration and high computer performance, and are not efficient in achieving accurate recognition of virtual elements without direct geometric relationships.
Innovation Solution
A method that determines the intersection of viewing and indicating directional vectors with a target plane, using tolerance ranges and correction factors to associate intersection coordinates with target objects, allowing for accurate recognition without initial calibration, and enabling activation of target recognition only when needed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If 3-D image recording devices and spatial relationship analysis are used to improve target accuracy, then measurement precision is improved, but device complexity and computer performance requirements increase
Solution Approach 1:
The patent extracts only the essential geometric relationships needed for target recognition - specifically the directional vectors from eye position and finger indication - and processes these extracted features to determine intersections with the target plane. This selective extraction of critical spatial information reduces computational complexity while maintaining measurement precision.
Solution Approach 2:
The patent creates a simplified geometric model (copy) of the spatial relationship between the user's eye, finger, and target plane. Instead of processing complete 3-D image data, it uses copied directional vectors and intersection calculations that replicate the essential spatial information in a computationally efficient manner.
2Measurement precision
If calibration procedures are implemented to improve target recognition accuracy, then measurement precision is improved, but loss of time and ease of operation deteriorate
Solution Approach 1:
The system performs self-calibration by automatically determining the user's eye position and finger indication vectors through sensor data, then calculating the intersections with the target plane without requiring manual calibration procedures. The system serves itself by adapting to each user's unique spatial relationships in real-time.
Solution Approach 2:
The patent establishes the geometric model and intersection calculation framework in advance, so that when target recognition is needed, the system can immediately process the directional vectors and determine intersections without requiring preliminary calibration steps. The methodology is prepared beforehand to enable rapid, calibration-free operation.
3Speed
If continuous target recognition is performed to improve responsiveness, then speed is improved, but use of energy and device complexity increase
Solution Approach 1:
Instead of continuous target recognition, the system uses periodic or event-triggered recognition based on detected finger gestures. When a finger indication is detected, the system periodically calculates the intersection with the target plane to determine the target. This periodic action maintains responsiveness to user input while significantly reducing energy consumption compared to continuous processing.
Data Source
AI summary
A method for the target recognition of target objects, in particular for the target recognition of the operating elements in a motor vehicle are disclosed herein. The method (and systems which incorporate the method) may employ various gaze tracking device, hand tracking devices, and combine inputs from the two based on a methodology for correlation.