AR HMD Scene Camera Automatic Calibration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing augmented reality (AR) devices with optical see-through head-mounted displays require cumbersome calibration processes, typically involving 12 steps per eye, and user-specific calibration data, making it difficult for users to achieve well-aligned 3D augmented reality with real-world objects, as projection matrices vary between individuals.
Innovation Solution
An automatic calibration method that computes the projection matrix on the fly using eye gaze base point readings, real-world object coordinates, and transformations across different coordinate systems, eliminating the need for manual calibration and allowing for real-time adaptation to changes in the user's eye position and device placement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual calibration process is used for each user, then alignment precision of augmented reality with real world objects is improved, but device complexity and time consumption increase significantly
Solution Approach 1:
The system performs automatic calibration without user intervention by capturing images of calibration patterns, detecting feature points, and computing projection matrices through automated image processing algorithms. The calibration process serves itself by using the device's own camera and processor to complete the alignment task.
Solution Approach 2:
The patent replaces the manual mechanical calibration process with an automated optical-digital system. Instead of manual adjustment of physical components, the system uses the scene camera to capture images, processes them through computer vision algorithms to detect feature points, and computes calibration parameters digitally to achieve precise alignment.
2Measurement precision
If user-specific calibration data is collected, then alignment precision for individual users is improved, but device complexity and data management burden increase
Solution Approach 1:
The patent develops a universal calibration approach that works for all users without requiring individual calibration data collection. The same automated calibration process and projection matrix computation method can be applied to any user, making the system universally applicable while reducing data management complexity.
Solution Approach 2:
The system uses a standardized calibration pattern (such as a chessboard or circular markers) that can be copied and used repeatedly for calibration. This standardizable calibration target allows the same calibration process to be applied universally across different users and devices without requiring custom calibration data for each individual.
3Manufacturing precision
If 12-step calibration process is performed for each eye, then augmented reality alignment is improved, but ease of operation deteriorates
Solution Approach 1:
The patent merges the calibration processes for both eyes into a single automated workflow. The system captures calibration images, processes them through unified image processing algorithms, and computes projection matrices for both eyes simultaneously, reducing the操作步骤 from 12 separate steps to a single automated operation.
Solution Approach 2:
The manual 12-step calibration process is replaced with an automated computer vision system that captures images and computes calibration parameters through digital processing. This substitution eliminates the need for users to manually perform multiple calibration steps while maintaining high alignment precision.
Data Source
AI summary
An apparatus for calibrating an augmented reality (AR) device having an optical see-through head mounted display (HMD) obtains eye coordinates in an eye coordinate system corresponding to a location of an eye of a user of the AR device, and obtains object coordinates in a world coordinate system corresponding to a location of a real-world object in the field of view of the AR device, as captured by a scene camera having a scene camera coordinate system. The apparatus calculates screen coordinates in a screen coordinate system corresponding to a display point on the HMD, where the calculating is based on the obtained eye coordinates and the obtained object coordinates. The apparatus calculates calibration data based on the screen coordinates, the object coordinates and a transformation from the target coordinate system to the scene camera coordinate system. The apparatus then derives subsequent screen coordinates for the display of AR in relation to other real-world object points based on the calibration data.