Datum Plane Determination for AR Scene Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing optical see-through Augmented Reality (AR) systems require manual adjustment of lens angles or calibration positions to match virtual objects with real scenes, leading to suboptimal real-time alignment and user experience.
Innovation Solution
A method and system for determining a datum plane by acquiring a depth image, performing edge extraction to form edge images with planar graphs, and selecting the appropriate planar graphs to establish a datum plane, allowing for the creation of a virtual coordinate system that aligns virtual objects with real scenes in real time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual adjustment of lens angles or calibration positions is used to match virtual objects with real scenes, then alignment accuracy can be improved, but device complexity and ease of operation deteriorate
Solution Approach 1:
The system automatically detects planar graphs in the real scene and determines the datum plane without requiring manual adjustment of lens angles or calibration positions. The AR device performs self-calibration by processing depth images and edge extraction algorithms to identify horizontal planes, thereby achieving automatic alignment between virtual objects and real scenes.
Solution Approach 2:
The system changes the operational parameters from manual lens angle adjustment to automatic depth image processing. By transforming the input data format (from manual calibration inputs to automated depth image analysis) and processing parameters (edge extraction, planar graph detection), the system achieves both high alignment accuracy and ease of operation.
2Measurement precision
If manual calibration is performed to achieve proper positioning of virtual objects, then positioning accuracy improves, but time consumption increases
Solution Approach 1:
The system performs preliminary action by continuously analyzing the real scene environment and pre-determining the datum plane before virtual object placement is required. The depth image acquisition and edge extraction processes are continuously running, so when virtual object positioning is needed, the system already has the calibrated reference plane ready, eliminating the need for time-consuming manual calibration at the moment of use.
Solution Approach 2:
The system maintains continuous useful action by continuously acquiring depth images and performing edge extraction to track the real scene environment. This continuous processing ensures that the datum plane determination is always up-to-date, allowing immediate and accurate virtual object positioning without interrupting the workflow for calibration.
3Adaptability or versatility
If real-time matching of virtual objects with real scenes is achieved, then user experience improves, but device complexity increases
Solution Approach 1:
The system segments the complex task of real-time AR alignment into distinct processing stages: depth image acquisition, edge extraction to form edge images, planar graph detection from edge images, and datum plane determination. By dividing the processing pipeline into modular segments, the system achieves real-time adaptation capability while managing device complexity through structured, incremental processing.
Data Source
AI summary
A method and a system for determining a datum plane are disclosed. The method for determining a datum plane includes: acquiring a depth image; performing edge extraction on the depth image to form an edge image, the edge image including a plurality of planar graphs; and selecting from the planar graphs in the edge image to determine the datum plane. The technical solutions provided by the disclosure can easily match the virtual object with the real scene in real time, and improve users' sensory experience beyond reality.