AR Headset Marker Positioning via Coordinate Transformation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current augmented reality techniques based on visual marker scanning are unsatisfactory when the marker is at a great distance from the operator or when the virtual information is far from the marker, leading to errors in orientation and positioning of displayed virtual information.
Innovation Solution
A method implemented by a computer in an augmented reality headset that receives images of an environment, calculates the distance and viewing angle of a visual marker, corrects the viewing angle based on features of the support plane, and determines display features for augmented reality elements, ensuring accurate positioning and orientation regardless of the user's position relative to the marker.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If visual marker scanning is used to display virtual information, then the implementation is simple and direct, but the positioning precision deteriorates when the marker is at a great distance or the virtual information is far from the marker
Solution Approach 1:
The patent introduces an intermediary coordinate transformation system that mediates between the visual marker coordinate system and the display coordinate system. By establishing transformation matrices and performing multi-step coordinate conversions, the system accurately determines the position and orientation of virtual information even when the marker is distant or the virtual content is far from the marker, thus resolving the positioning precision issue while maintaining implementation feasibility
Solution Approach 2:
The patent transitions from two-dimensional marker image processing to three-dimensional spatial coordinate transformation. By introducing depth information and performing 3D coordinate system transformations, the system accurately calculates the position and orientation of virtual information in space, overcoming the limitations of 2D marker scanning for distant objects
2Area of stationary object
If the visual marker is placed at a great distance from the operator, then the field of view coverage is improved, but the orientation precision deteriorates due to angle errors
Solution Approach 1:
The patent replaces traditional mechanical angle measurement methods with computer vision-based coordinate transformation. By using image processing to detect marker features and performing mathematical coordinate transformations, the system accurately determines orientation without being affected by distance-related angle errors, thus maintaining precision while expanding field of view coverage
3Adaptability or versatility
If the virtual information is displayed far from the visual marker, then the spatial distribution is improved, but the positioning accuracy deteriorates due to error amplification
Solution Approach 1:
The patent performs preliminary coordinate transformation calculations to establish the relationship between the marker coordinate system and the display coordinate system before actual virtual information rendering. By pre-calculating transformation matrices and storing them for later use, the system efficiently and accurately determines the position of virtual information even when displayed far from the marker, avoiding error amplification while maintaining spatial distribution flexibility
Data Source
AI summary
The present invention relates to a method for determining display features of an augmented reality element, the method comprising the steps of receiving an image depicting a visual marker on a support surface, obtaining correspondence data between the augmented reality element and the visual marker, calculating the distance and a viewing angle relative to the visual marker and the user, correcting the viewing angle calculated as a function of features relating to a support plane containing the support surface, to obtain a corrected viewing angle, and determining display features of the augmented reality element based on the correspondence data, a rotation of the visual marker relative to a reference, the calculated distance and the corrected viewing angle.