AR Surface Detection via Horizontal Anchor and Intersection Geometry
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional approaches to rendering augmented reality (AR) content struggle with optically challenging surfaces such as bare walls, glass panels, or reflective or textureless surfaces, which lack distinct features detectable by optical sensors, leading to inaccurate positioning and tracking of AR content.
Innovation Solution
The method involves detecting and tracking horizontal surfaces to establish an anchor point, then using user input or computer vision techniques to determine the intersection with vertical surfaces, generating a virtual plane locked to the surface's orientation, allowing AR content to be accurately positioned and rendered on optically challenging surfaces like walls.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional optical sensor approaches are used to detect surfaces, then detection is straightforward on surfaces with distinct features, but detection fails on optically challenging surfaces that lack distinct features
Solution Approach 1:
The patent introduces horizontal surfaces as intermediary reference objects to indirectly detect vertical optically challenging surfaces. Instead of directly detecting the featureless vertical wall, the system detects the horizontal floor surface and uses their geometric intersection relationship to infer the position and orientation of the vertical surface, solving the detection problem for surfaces lacking distinct features
Solution Approach 2:
The patent replaces direct optical feature detection with a geometric relationship-based detection system. By substituting the mechanical/optical direct detection approach with a mathematical model that calculates vertical surface properties from horizontal surface detection data and intersection geometry, the system achieves detection capability on optically challenging surfaces
2Reliability
If direct detection of vertical surfaces is attempted, then positioning can be achieved on surfaces with features, but tracking becomes unreliable on featureless surfaces
Solution Approach 1:
The patent performs preliminary detection and tracking of horizontal surfaces before using them as references for vertical surface positioning. By first establishing a reliable horizontal surface anchor and calculating its intersection with the vertical surface, the system creates a stable reference framework that enables subsequent reliable tracking of vertical surfaces without requiring features on those surfaces
Solution Approach 2:
The horizontal surface acts as an intermediary reference that transfers positioning reliability from the easily detectable horizontal plane to the difficult-to-detect vertical plane. The system tracks the horizontal surface with high reliability and uses the geometric intersection relationship to maintain reliable positioning on the vertical surface, bypassing the need for vertical surface features
3Manufacturing precision
If conventional AR rendering is used on horizontal surfaces, then acceptable results are achieved, but rendering on vertical surfaces lacks accuracy
Solution Approach 1:
The patent extends the AR rendering capability from the horizontal plane to the vertical dimension by calculating the intersection of horizontal and vertical surfaces. This dimensional extension allows the system to project AR content onto vertical surfaces with the same precision as horizontal surfaces, achieving accurate placement in three-dimensional space across different surface orientations
Solution Approach 2:
The patent creates a universal surface detection and rendering system that works equally well on both horizontal and vertical surfaces. By using the intersection of horizontal reference surfaces with vertical target surfaces, the system achieves consistent rendering accuracy across different surface orientations, making the AR system adaptable to multiple surface types and orientations
Data Source
AI summary
Approaches in accordance with various embodiments provide for the presentation of augmented reality (AR) content with respect to optically challenging surfaces. Such surfaces can be difficult to locate using conventional optical-based approaches that rely on visible features. Embodiments can utilize the fact that horizontal surfaces can be located relatively easily, and can determine intersections or boundaries of those horizontal surfaces that likely indicate the presence of another surface, such as a vertical wall. This boundary can be determined automatically, through user input, or using a combination of such approaches. Once such an intersection is located, a virtual plane can be determined whose relative location to a device displaying AR content can be tracked and used as a reference for displaying AR content.