Landmark Database Pruning for AR Tracking Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image-based localization and tracking systems in augmented and virtual reality face challenges in maintaining accurate object tracking and localization due to outdated landmark databases, leading to inefficient computation, power consumption, and reduced accuracy, especially on mobile devices.
Innovation Solution
A tracking system that utilizes a deep-learning model to label landmarks with semantic information, updates the landmark database by detecting and labeling landmarks from real data capture, and employs a map health system to delete stale landmarks, ensuring only relevant data is maintained for efficient and accurate tracking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the landmark database is continuously updated with all detected landmarks, then the completeness of the database is improved, but the computational intensity and power consumption increase
Solution Approach 1:
The system selectively removes stale landmarks from the database that are no longer relevant to current tracking needs, while retaining important landmarks. This reduces the database size and computational load for mobile devices while maintaining tracking accuracy for objects of interest
Solution Approach 2:
The system extracts and removes unnecessary stale landmarks from the database, separating useful tracking data from obsolete information. This extraction process reduces computational intensity without compromising the accuracy of current object tracking
2Reliability
If the landmark database is continuously updated with all detected landmarks, then the completeness of the database is improved, but the device complexity increases
Solution Approach 1:
The system automatically discards stale landmarks that are no longer relevant, reducing database complexity. This automated pruning mechanism simplifies the overall system while maintaining tracking reliability through selective data retention
3Use of energy by moving object
If the landmark database is not updated, then the computational intensity is reduced, but the accuracy of localization and tracking deteriorates
Solution Approach 1:
The system updates the database by adding new landmarks while simultaneously removing stale ones, maintaining an optimal balance between database freshness and computational efficiency. This selective update process preserves localization accuracy without excessive power consumption
Data Source
AI summary
In one embodiment, a computing system is configured to, during a first tracking session, detect first landmarks in a first image of the environment surrounding a user, and determine a first location of the user by comparing detected first landmarks to a landmark database. During a second tracking session, the computing system captures motion data and estimates a second location of the user based on the motion data and first user location. Based on the motion data and first user location, the computing system detects landmarks in a second image at a second location. The system accesses expected landmarks from the landmark database visible at the second location and determines the estimated second location of the user is inaccurate by comparing the expected landmarks with the second landmarks. The computing system re-localizes the user by comparing the landmarks in the landmark database and third landmarks in a third image.