AR Marker Clustering and Representative Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current augmented reality (AR) technologies face challenges in efficiently generating and managing markers and AR objects, as well as accurately recognizing markers in real-world scenes, which hinders the seamless integration of virtual information with real-world environments.
Innovation Solution
The method involves clustering markers based on attributes, determining representative markers, and using marker correction information to enhance marker recognition and AR object placement, allowing for efficient and accurate overlay of AR content on real-world scenes by prioritizing marker searches and utilizing user-provided data to correct and match virtual objects with real-world positions and orientations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If markers are clustered and representative markers are determined to improve marker identification efficiency, then marker recognition speed increases, but system complexity increases due to clustering algorithms and representative marker selection
Solution Approach 1:
The patent segments the large set of all markers into multiple clusters based on attributes such as location, type, or other characteristics. Each cluster is then represented by a representative marker, transforming a single large search space into multiple smaller, organized groups. This segmentation reduces the time required to identify relevant markers while managing system complexity through structured organization.
Solution Approach 2:
The patent introduces representative markers as intermediary elements between the full marker set and the actual marker identification process. These representative markers serve as proxies or mediators that represent entire clusters, allowing the system to quickly filter and search through clustered markers without examining every individual marker, thus improving identification efficiency.
2Productivity
If preferential searching of clustered markers is implemented to improve recognition speed, then marker search efficiency increases, but computational overhead increases due to clustering maintenance and representative marker management
Solution Approach 1:
The patent performs clustering and representative marker determination in advance, before the actual marker identification process. By pre-organizing markers into clusters and selecting representative markers beforehand, the system reduces the computational burden during real-time operation. The preferential searching of clustered markers then becomes a simpler, faster process that leverages the pre-computed structure.
3Measurement precision
If marker correction information is used to improve marker recognition accuracy, then positioning precision increases, but device complexity increases due to additional correction mechanisms
Solution Approach 1:
The patent implements a feedback mechanism where marker correction information is obtained based on the recognized scene and used to adjust and refine marker identification. The system continuously compares the recognized scene with stored marker information, obtains correction data, and applies it to improve recognition accuracy. This feedback loop enhances precision while managing complexity through iterative refinement.
4Measurement precision
If comprehensive marker attributes are used for clustering to improve recognition accuracy, then marker identification precision increases, but data processing complexity increases
Solution Approach 1:
The patent applies different clustering strategies and attributes based on local requirements and contexts. Different clusters may use different attributes for grouping (such as location-based clustering for spatial markers or type-based clustering for functional markers). This local quality approach allows the system to improve identification accuracy for specific marker types while managing overall data processing complexity through targeted, context-specific processing.
Data Source
AI summary
An augmented reality (AR) service clustering a plurality of markers for mapping an AR object by a device into at least one group and determining a representative marker of the clustered group, and preferentially searching for markers included in a cluster of the representative marker when a scene recognized by the device corresponds to the representative marker, is provided. An AR service generating an AR object based on data received from a user while a scene recognized by a device is displayed on a screen of the device and determining the recognized scene as a marker of the AR object is also provided. An AR service clustering a previously obtained plurality of pieces of content based on a predetermined reference and generating an AR object based on the plurality of clustered pieces of content is further provided.