AR Plane Detection via Edge Filtering and Line Approximation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing augmented reality (AR) systems face challenges in accurately tracking real-world objects with low-texture surfaces and overlapping AR content with real-world objects, leading to interference and accuracy issues, particularly in markerless tracking and facade plane detection.
Innovation Solution
A computer-implemented method using adaptive binarization, variance thresholding, and Hough transform techniques to detect edges, lines, and iteratively approximate distributions for determining the spatial configuration of objects, enabling markerless tracking and accurate placement of AR content on planar and curved surfaces without additional markers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If hand-drawn markers are used for markerless tracking, then tracking accuracy is improved, but the complexity of the system increases due to the need to draw or print additional elements
Solution Approach 1:
The system uses the existing texture patterns on the object surface itself for tracking, without requiring external markers. The algorithm extracts and tracks natural texture features that are already present on the object, making the object self-sufficient for tracking purposes.
Solution Approach 2:
The invention extracts tracking features directly from the existing texture of the object surface, separating the tracking function from the object's primary function. This allows the object to be tracked without adding external markers, by taking out and utilizing only the relevant texture information.
2Loss of information
If AR content is displayed to provide useful information, then information delivery is improved, but the view of real-world objects is obscured
Solution Approach 1:
The AR content is selectively positioned in specific regions of the display based on the detected object boundaries and important real-world features. Different regions of the display have different qualities - some areas show AR information while others preserve clear views of real-world objects, optimizing both information delivery and visibility.
Solution Approach 2:
The system transitions from two-dimensional AR content placement to three-dimensional spatial awareness by detecting object boundaries and depths. AR content is positioned in relation to the detected object geometry, creating a multi-layered display where virtual and real elements coexist in a spatially-aware composition.
3Adaptability or versatility
If template update approach is used for tracking, then adaptability to transforms is improved, but patches drift occurs due to slow descriptor changes
Solution Approach 1:
The system implements periodic template updates at controlled intervals rather than continuous updates. This periodic action allows the descriptor to adapt to transforms over time while maintaining stability between updates, preventing both drift and excessive reactivity to temporary variations.
Solution Approach 2:
The tracking system uses feedback from successful trackings to selectively update templates only when confidence thresholds are met. This feedback mechanism ensures that updates occur only when the transformation is significant and reliable, maintaining stability while adapting to genuine changes in object appearance.
Data Source
AI summary
A computer-implemented method of using augmented reality (AR) to detect a plane and a spatial configuration of an object, the method comprising the steps of detecting one or more edges of the object; identifying one or more lines of the object; filtering the one or more lines of the object; iteratively approximating one or more distributions of the one or more lines; and determining boundaries of the object.