AR Point Cloud Tiling for Large-Scale Location Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current AR systems face challenges in generating and efficiently displaying large-scale point clouds due to the computational demands and logistical difficulties in data collection, making it impractical for use in mobile devices.
Innovation Solution
An AR system that accesses data objects with image and location data, applies transformations to generate a point cloud, assigns it to a location, and loads a portion of the point cloud to a client device based on contextual conditions such as location, time, device attributes, and network connectivity, optimizing computational demands.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If a complete point cloud is generated for large-scale environments, then AR content coverage is improved, but computational burden and data transmission requirements increase significantly
Solution Approach 1:
The patent divides the complete point cloud into multiple tile-based portions, where each tile represents a specific geographic region. The system generates and transmits only the tile portions that are relevant to the user's current location and field of view, rather than transmitting the entire point cloud. This segmentation approach enables AR content coverage across large-scale environments while significantly reducing computational burden and data transmission requirements for mobile devices.
2Area of stationary object
If a complete point cloud is generated for large-scale environments, then AR content coverage is improved, but data transmission requirements increase significantly
Solution Approach 1:
The patent segments the point cloud data into tile-based portions corresponding to different geographic regions. The system determines which tile portions are visible to the user based on their location and orientation, and transmits only those specific portions. This selective transmission of segmented data enables coverage of large-scale environments while minimizing the volume of data that needs to be transmitted over the network.
3Speed
If point cloud data is pre-loaded to a mobile device, then AR display speed is improved, but device memory requirements and initial load time increase
Solution Approach 1:
The system performs preliminary processing of the point cloud data on the server side by organizing it into tile-based portions and pre-computing visibility information. When a user requests AR content, the system quickly identifies and transmits only the specific tile portions that are currently visible to the user. This preliminary organization and selective transmission approach enables fast AR display performance without requiring the mobile device to store large amounts of point cloud data in memory.
Data Source
AI summary
An Augmented-Reality which performs operations that include: accessing a data object that comprises image data, location data, and orientation data; applying a transformation to the data object to produce a rectified data object; generating a point cloud based on the rectified data object; assigning the point cloud to a location based on at least the location data of the data object; detecting a client device at the location; and loading the point cloud to the client device in response to the detecting the client device at the location.