Feature Point Map Management for AR Camera Localization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In augmented reality, the accurate location and orientation of a user device's camera are often degraded due to the extraction of feature points from images that include mobile objects, which are not related to the pre-existing feature point map, leading to errors and reduced user satisfaction.
Innovation Solution
A method and apparatus for managing a feature point map that involves generating an initial travel path for an environmental information collecting entity, detecting mobile objects, modifying the path if occlusion occurs, and updating the feature point map based on changes in feature point information to ensure accurate location and orientation estimation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If feature points are extracted from all objects in the camera image without considering mobile objects, then the feature point extraction process is simple and fast, but the accuracy of camera location and orientation estimation deteriorates due to errors from transient objects
Solution Approach 1:
The patent segments the feature point extraction process into two distinct phases: initial feature point map generation using all detectable features, and subsequent verification phase where mobile objects are detected and excluded from feature matching. This segmentation allows the system to maintain simple initial extraction while improving accuracy through targeted mobile object detection and exclusion in the verification stage.
Solution Approach 2:
The patent performs preliminary mobile object detection and exclusion before feature point matching. By detecting mobile objects in advance and generating exclusion masks, the system prevents transient objects from contaminating the feature point extraction process, thereby improving camera localization accuracy without requiring complex real-time filtering during matching.
2Adaptability or versatility
If the feature point map is updated frequently to account for environmental changes, then the adaptability to changing surroundings is improved, but the computational overhead and processing time increase
Solution Approach 1:
The patent implements periodic verification of the feature point map at predetermined time intervals rather than continuous updates. During each verification period, mobile objects are detected and the feature point map is updated if necessary. This periodic approach maintains adaptability to environmental changes while significantly reducing computational overhead compared to continuous update mechanisms.
Solution Approach 2:
The system incorporates feedback mechanisms where detection results from mobile object detection are fed back to determine whether feature point map updates are necessary. By using feedback to conditionally trigger updates only when environmental changes are detected, the system maintains high adaptability while minimizing unnecessary processing time and computational resources.
Data Source
AI summary
Provided are an apparatus and method for managing a feature point map. The method includes generating an initial travel path through which an environmental information collecting entity moves, determining the generated initial travel path as a travel path for detection and controlling the environmental information collecting entity along the determined travel path for detection, receiving environmental information from the environmental information collecting entity, extracting feature point information from the environmental information, generating a feature point map from the feature point information, and storing the feature point map.