Landmark Based Positioning Using Virtual 2D Depth Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing navigation systems face challenges in indoor environments where satellite signals are unreliable, and methods like ray tracing in three dimensions are computationally complex, especially in multi-level venues with landmarks on different elevations.
Innovation Solution
Converting a two-dimensional map into a three-dimensional model with color-coded areas where a mobile imaging device can potentially be located, such as floors and escalators, and using pose parameters to render a selected view into a virtual two-dimensional image, allowing for location estimation by matching landmarks and pose parameters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If ray tracing in three dimensions is used for location estimation, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent converts a three-dimensional venue map into a two-dimensional depth image representation. By projecting 3D spatial information onto a 2D plane with depth encoding, the system simplifies computational requirements while preserving location estimation accuracy. The depth image maintains spatial relationships necessary for landmark matching without requiring complex 3D ray tracing operations.
2Measurement precision
If three-dimensional maps with color-coded areas are used, then measurement precision is improved, but use of energy increases
Solution Approach 1:
The patent creates a simplified two-dimensional depth image copy of the three-dimensional venue map. This 2D representation serves as a computationally efficient substitute for processing full 3D spatial data. The depth image encodes spatial information in a format that requires less computational energy while maintaining the precision needed for accurate location estimation through landmark matching.
Data Source
AI summary
Disclosed are devices, methods and storage media for use in determining position information for imaging devices or mobile devices. In some implementations, a landmark is identified in an image which is obtained from an imaging device which in turn is positioned at a location and in a pose. A virtual two-dimensional image that would be visible from the landmark is determined based, at least in part, on the pose. The location or position information is based, at least in part, on the virtual two-dimensional image.