Augmented Reality Content Alignment Using Stereo Vision
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing augmented reality (AR) devices face challenges in accurately aligning virtual content with real-world objects due to computational limitations and the need for specialized depth sensors, which are not feasible in compact AR devices.
Innovation Solution
Utilizing stereo vision with commonly available color cameras to determine the surface plane and edges of real-world objects, allowing for precise alignment of virtual content relative to these features using edge alignment techniques.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If LIDAR sensors are used to estimate surface characteristics, then measurement precision is improved, but device complexity and cost increase
Solution Approach 1:
The patent uses color camera images (2D copies) to represent and analyze real-world surfaces, replacing the need for direct 3D measurement with LIDAR. By processing 2D image data to infer 3D surface properties, the system achieves accurate surface estimation without specialized depth sensors.
Solution Approach 2:
The patent replaces the mechanical/optical LIDAR measurement system with a computational imaging approach using standard color cameras. Instead of actively emitting light and measuring time-of-flight, the system passively captures color images and uses algorithms to extract depth and surface information.
2Measurement precision
If LIDAR sensors are used for depth estimation, then measurement precision is improved, but productivity decreases due to computational expense
Solution Approach 1:
The patent extracts only the essential information needed for surface estimation from color images, rather than performing full 3D reconstruction. By focusing on extracting depth and surface normal information from 2D images, the system reduces computational overhead while maintaining accuracy.
Solution Approach 2:
The patent performs partial depth estimation by analyzing only relevant regions of images and using simplified geometric models. Instead of computing complete 3D point clouds, the system estimates surface properties at key locations, reducing processing requirements while maintaining sufficient accuracy for AR applications.
3Measurement precision
If specialized depth sensors are used, then measurement precision is improved, but ease of manufacture worsens
Solution Approach 1:
The patent makes the color camera serve multiple functions: capturing 2D images for visual display, extracting depth information for 3D surface estimation, and providing texture information. This multi-functionality eliminates the need for separate depth sensors, simplifying device manufacturing and assembly.
Data Source
AI summary
A system is disclosed, including a processor and a memory. The memory stores instructions that, when executed by the processor, configure the system to perform operations. Surface plane information is obtained, defining a surface plane passing through a surface location and oriented according to a surface normal. An edge is detected in an image. Virtual content is presented, having a virtual position based on an orientation of the edge and the surface plane information.