Augmented Reality Marker Confidence Ranking for Indoor Positioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current augmented reality systems rely on GPS and other sensors, which are unreliable indoors and provide positional accuracy within only one to three feet, making it difficult to accurately determine the position of objects in images when they are close together.
Innovation Solution
A computer-implemented method that uses image data from a camera to detect multiple markers, assigns confidence levels to each marker, selects the marker with the highest confidence, and generates overlaid display data based on transformation and positional offset parameters to improve positional accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If GPS and sensor data are used to determine device position, then the system can provide location information, but the positional accuracy is limited to within one to three feet which is insufficient when objects are close together
Solution Approach 1:
The patent introduces markers as intermediary objects placed in the environment to serve as reference points for position determination. These markers act as mediators between the camera system and the objects of interest, enabling precise relative positioning by detecting marker positions and using them to calculate device location and orientation, thereby achieving accuracy much finer than GPS capabilities.
Solution Approach 2:
The patent replaces the GPS-based mechanical positioning system with an optical marker detection system. Instead of relying on satellite signals and sensor fusion, the system uses computer vision to detect markers in the camera field of view, transforming the positioning problem from a global coordinate system challenge into a local visual recognition and geometric calculation problem.
2Measurement precision
If multiple markers are detected in an image, then more positional information is available, but determining which marker to use becomes more complex
Solution Approach 1:
The patent implements a feedback mechanism where the system calculates confidence levels for each detected marker based on multiple factors including marker detection quality, geometric consistency, and positional plausibility. This feedback information is used to automatically select the most reliable marker, resolving the selection complexity through an objective, algorithm-driven process rather than manual intervention or simple heuristics.
Solution Approach 2:
The patent transforms the marker selection problem from a qualitative judgment into a quantitative comparison by introducing confidence level parameters. Each marker is evaluated across multiple parameters (detection confidence, geometric consistency, positional validity), and the marker with the highest composite confidence score is selected, converting a complex multi-criteria decision into a straightforward parameter comparison.
Data Source
AI summary
A computer implemented method for augmenting a display image includes receiving image data, the image data including data representing one or more objects, and at least a first marker and a second marker. The method includes receiving a first confidence level for the first marker and a second confidence level for the second marker. The method includes determining a selected marker from the first marker and the second marker. The selected marker is determined according to a highest confidence level of the first confidence level and the second confidence level. The method includes determining a transformation and a positional offset for the selected marker. The method includes generating overlaid display data for the one or more objects in the image data, the one or more objects determined in accordance with the transformation and the positional offset.