AR View Alignment Using Floorplan-Corrected Camera Pose
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing augmented reality applications suffer from alignment issues due to inaccurate camera pose estimates in unknown three-dimensional spaces, leading to inconsistencies and misalignments in composite views.
Innovation Solution
Utilizing a priori information about the physical environment, such as floorplans, to generate and correct camera pose estimates by minimizing a cost function that constrains walls to meet at right angles, resulting in a revised and more accurate floorplan estimate.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If camera pose estimates are generated in unknown three-dimensional spaces, then augmented reality views can be displayed, but alignment accuracy deteriorates due to errors in pose estimation
Solution Approach 1:
The system performs preliminary actions by using a priori information about the physical environment (such as floorplans, wall locations, and geometric constraints) to pre-establish a reference framework before displaying augmented reality views. This preliminary structuring of spatial information enables accurate pose estimation even in previously unknown spaces by providing constraints that guide the alignment process.
Solution Approach 2:
The system implements feedback mechanisms by continuously comparing the estimated camera pose with constraints derived from a priori environmental information. When misalignments are detected, the system adjusts the pose estimates to satisfy geometric constraints (such as walls meeting at right angles), creating a closed-loop correction process that improves alignment accuracy iteratively.
2Measurement precision
If a priori information about the physical environment is utilized to correct camera pose estimates, then alignment accuracy improves, but system complexity increases
Solution Approach 1:
The system changes parameters by transforming raw camera pose estimates into corrected poses that satisfy geometric constraints. By applying mathematical transformations that enforce known environmental properties (such as orthogonal wall intersections), the system improves accuracy without requiring complex hardware modifications, instead using computational geometry to adjust pose parameters.
Solution Approach 2:
The system uses simplified representations or copies of environmental information (such as 2D floorplans or constraint models) to correct 3D camera pose estimates. These simplified models serve as reference frameworks that are computationally efficient to process while providing sufficient geometric constraints to achieve accurate alignment, avoiding the need for complex full 3D environmental mapping.
Data Source
AI summary
Techniques for generating aligned, augmented reality views are disclosed. In some embodiments, an estimate of a floorplan of a three-dimensional space is first generated based on one or more received views of the three-dimensional space from a camera located in the three-dimensional space. A camera pose of the camera in the three-dimensional space is next determined based on generating a corrected floorplan from the estimate of the floorplan based on foreknowledge associated with the floorplan. Display of a rendering having the determined camera pose is facilitated in an actual view of the camera, resulting in an aligned, augmented reality view.