AR-HUD Alignment Using 2D Mapping for Moving Line of Sight
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing immersive augmented reality head-up displays (AR-HUDs) face challenges in maintaining precise virtual-real alignment due to driver head movements and vehicle motion, leading to reduced alignment precision and limited application scenarios.
Innovation Solution
A method and apparatus that determine a user's line-of-sight focusing area, extract entity identification points from photographed images, and project these points onto virtual images using two-dimensional mapping relationships, enabling alignment with real-world objects, without requiring three-dimensional reconstruction or complex hardware.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If three-dimensional calibration re-projection technology is used to re-project out-of-vehicle scenery to human eyes, then virtual-real alignment precision is improved, but sensor requirements and computing power requirements increase significantly
Solution Approach 1:
The patent extracts and utilizes only the essential two-dimensional mapping relationship between the camera image plane and the virtual image plane, eliminating the need for complex three-dimensional reconstruction. By focusing on the 2D homography transformation, the system achieves virtual-real alignment without requiring high-end sensors or substantial computing power.
Solution Approach 2:
The patent employs a cost-effective calibration method using a calibration board with identifiable features that can be easily captured by the camera. This approach replaces expensive multi-sensor systems with a simple, disposable calibration target that provides sufficient accuracy for the application.
2Device complexity
If two-dimensional alignment technology is used to directly paste virtual image content according to a fixed scenario, then device complexity is reduced, but adaptability to different scenarios and line-of-sight movements is limited
Solution Approach 1:
The patent implements a dynamic alignment system that continuously updates the two-dimensional mapping relationship based on real-time camera images and detected entity positions. Instead of using a fixed scenario, the system adapts to different scenarios and line-of-sight movements by recalculating the homography transformation parameters, enabling versatile application while maintaining low device complexity.
Solution Approach 2:
The patent creates a universal alignment method that works across different scenarios (ground plane, uphill, downhill, above-ground objects) by using entity identification and adaptive 2D mapping. The same basic algorithm can handle various scenarios by detecting different entity types and adjusting the mapping accordingly, making the system multi-functional without increasing complexity.
3Measurement precision
If the virtual image size is constrained to match a fixed depth scenario, then alignment precision for that scenario is improved, but the application scenario range is limited
Solution Approach 1:
The patent dynamically adjusts the virtual image parameters (size, position, orientation) based on the detected entity distance and scenario type. By changing these parameters adaptively rather than fixing them for a single depth, the system maintains alignment precision across diverse scenarios including ground plane, elevated objects, and varying distances.
Data Source
Figure 1~2
Figure 3~5
Figure 6~8
AI summary
The present application discloses an alignment method and alignment apparatus for a display device, and a vehicle-mounted display system, and relates to the field of image processing. The alignment method includes: determining a line-of-sight focusing area of a user in a target carrier; acquiring a photographed image collected by a first camera for an original scenario, and extracting an entity identification point from the photographed image; determining a two-dimensional image mapping relationship between the photographed image and a virtual image in the current field of view of the user by means of pre-obtained calibration parameters, wherein the virtual image is an image to be projected by the display device; and projecting the entity identification point into the virtual image according to the two-dimensional image mapping relationship, wherein a virtual marker in the virtual image is aligned with an entity reference object in the line-of-sight focusing area. The present application solves the technical problem in related arts that a driver and a passenger cannot obtain good virtual-real alignment "immersive" augmented reality experience while moving the line of sight, thus reducing the use interest of the driver and the passenger.