Mixed Reality Position Tracking Using Frame Index Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining the position and orientation of an image sensing device in mixed reality environments face challenges in maintaining consistency between frames, particularly when there are abrupt movements or changes in the indices used for measurement, leading to discontinuous and inaccurate results.
Innovation Solution
A method that includes detecting indices in images, calculating position and orientation using current and past frame data, and applying correction values to maintain consistency and account for index appearance and disappearance between frames, ensuring smooth tracking of the device's position and orientation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If nonlinear optimization method is used to calculate position and orientation for each frame, then measurement precision is improved, but calculation time increases and continuity between frames is lost
Solution Approach 1:
The patent performs preliminary calculation of position and orientation using all detected indices in each frame, then uses this result as the initial value for the next frame's calculation. This preliminary action ensures continuity between frames while maintaining measurement precision through the optimization process.
Solution Approach 2:
The patent implements feedback by using the position and orientation calculated in the current frame as the initial value for the next frame's calculation. This feedback mechanism maintains temporal continuity and ensures that the optimization process builds upon previous results, reducing calculation time while preserving precision.
2Measurement precision
If position and orientation are calculated independently for each frame, then measurement precision is improved, but consistency between frames deteriorates
Solution Approach 1:
The patent calculates position and orientation for each frame independently first, then uses this preliminary result as the initial value for the next frame's calculation. This ensures both measurement precision through independent optimization and consistency between frames through the continuity of initial values.
Solution Approach 2:
The patent implements feedback by using the current frame's calculated position and orientation as the initial value for the next frame. This feedback loop maintains consistency between frames while preserving the precision of independent frame-by-frame calculations.
3Adaptability or versatility
If all detected indices are used for position and orientation calculation, then adaptability is improved, but measurement precision deteriorates due to index appearance and disappearance
Solution Approach 1:
The patent preliminarily calculates position and orientation using all detected indices, then adjusts the calculation by considering only indices that are present in both the current and previous frames. This preliminary action maintains adaptability to index changes while improving measurement precision by excluding unstable indices.
Solution Approach 2:
The patent extracts and removes indices that appear or disappear between frames from the calculation set. By taking out these unstable indices, the patent maintains measurement precision while preserving adaptability to handle index changes in the environment.
Data Source
AI summary
The first values of position and orientation of an image sensing device are obtained by using positions of all indices in a frame image. Indices having the same inherent information between a first index group (FIG) in the frame image and a second index group (SIG) in a preceding frame image are detected as a common index. If the information of the SIG contains information different from that of the FIG, second values of the position and orientation are obtained by using the positions of the common indices in the preceding frame image. If the information of the FIG contains information different from that of the SIG, third values of the position and orientation are obtained by using the positions of the common indices in the frame image. The first values are corrected by using correction values based on the obtained first to third values.