Multi-Person AR Scene Matching for Shared Virtual Object Editing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In multi-person augmented reality (AR) systems, users are burdened with the need to input room ID numbers to access virtual object information, increasing operational complexity and memory load, especially when multiple AR scenes are involved.
Innovation Solution
A system architecture involving a cloud-based approach where a first terminal captures frame images, extracts image parameters, and sends them to a cloud server for determining corresponding virtual object information, which is then displayed and editable on the terminal, with editing results stored back in the cloud.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If users manually input room ID numbers to access virtual object information, then virtual object sharing can be realized, but user operational complexity increases and memory burden increases
Solution Approach 1:
The system automatically identifies the current AR scene and retrieves corresponding virtual object information without requiring user input of room ID numbers. The terminal device performs self-service by autonomously obtaining scene identification data and matching it with stored virtual object information, thereby eliminating manual operations while maintaining virtual object sharing functionality
Solution Approach 2:
The patent introduces an automatic scene recognition mechanism as an intermediary between the user and the virtual object information system. This intermediary automatically captures the current AR scene, extracts identification data, and retrieves corresponding virtual object information from the cloud server, thereby mediating the interaction and eliminating the need for manual room ID input
2Adaptability or versatility
If multiple AR scenes are supported, then scene versatility improves, but user memory burden increases
Solution Approach 1:
The system stores virtual object information in the cloud server rather than requiring users to memorize room ID numbers for multiple scenes. Each AR scene has its virtual object information copied and stored centrally, allowing users to access any scene's virtual objects without retaining memory of identification data, thereby supporting multiple scenes while minimizing user memory burden
Solution Approach 2:
The patent transitions the storage dimension from user memory to cloud-based database. Instead of users holding room ID numbers in their memory, the system uses a cloud server to store and manage virtual object information across multiple AR scenes, effectively moving the information burden from the user's cognitive dimension to an external storage dimension
Data Source
Figure 1~2
Figure 3~5
Figure 6~8
AI summary
An augmented reality (AR) processing method, an AR processing apparatus, an AR processing system, a computer readable storage medium, and an electronic device, relating to the technical field of AR. The AR processing method comprises: obtaining the current frame image acquired by a photographing module of a first terminal, extracting an image parameter of the current frame image, and sending the image parameter to the cloud, so that the cloud determines, using a pre-stored mapping result, information of a virtual object corresponding to the image parameter (S32); receiving the information of the virtual object sent by the cloud and displaying the virtual object (S34); and editing the virtual object in response to an editing operation for the virtual object (S36). Thus, operations of a user are reduced during a multi-person AR process, and the fun of multi-person AR experience can be enhanced.