Point Cloud Tile Tracks for Selective Geometry Retrieval
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing point cloud compression standards force users to download and decode all geometry-based point cloud component information, even when they are interested in specific regions or objects, lacking efficient partial access support.
Innovation Solution
A method and apparatus that utilize timed-metadata tracks to identify point cloud tiles within a scene, allowing for the retrieval and processing of specific geometry tile tracks corresponding to these tiles, enabling efficient partial access and rendering of point clouds.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all geometry-based point cloud component information is downloaded and decoded, then complete point cloud reconstruction is achieved, but data transfer volume and processing time increase unnecessarily
Solution Approach 1:
The point cloud data is divided into multiple tiles, each representing a specific spatial region. The encoded bitstream contains tile identification information that allows the decoding device to segment and process only the tiles corresponding to regions of interest, rather than decoding the entire point cloud data set.
Solution Approach 2:
The patent extracts and utilizes tile identification information from the encoded bitstream to selectively retrieve only the necessary tile data. This extraction mechanism enables the system to take out and process only the specific point cloud components needed for rendering particular spatial regions, eliminating unnecessary data transfer.
2Reliability
If all geometry-based point cloud component information is decoded, then complete point cloud rendering is achieved, but processing time and computational resources increase
Solution Approach 1:
The decoding process is segmented by tile, allowing the system to decode and render only the specific tiles corresponding to regions of interest. This segmentation reduces the total processing time while maintaining rendering quality for the desired spatial regions.
Solution Approach 2:
The system performs partial decoding action by processing only the necessary tiles rather than the complete point cloud data set. This partial action approach reduces computational overhead and processing time while still achieving the desired rendering outcome for specific spatial regions.
3Productivity
If selective access to specific point cloud regions is enabled, then data transfer efficiency improves, but system complexity increases due to tile tracking mechanisms
Solution Approach 1:
The patent introduces tile identification information as an intermediary element within the encoded bitstream. This intermediary contains spatial region metadata that mediates between the compressed point cloud data and the decoding process, enabling selective tile retrieval without requiring complex external tracking systems.
Solution Approach 2:
The tile identification information is merged into the encoded bitstream structure itself, combining the spatial region metadata with the compressed point cloud data. This merging eliminates the need for separate tracking mechanisms, reducing system complexity while maintaining selective access capability.
Data Source
AI summary
A method and apparatus include receiving a timed-metadata track identifying point cloud tiles corresponding to one or more spatial regions within a point cloud scene. A decoding device determines one or more point cloud tiles to be used for rendering an image. One or more geometry tile tracks are retrieved, via a communications network, corresponding to the determined one or more point cloud tiles. Each geometry tile track comprises point cloud geometry data for a respective tile. The retrieved geometry tile tracks are processed.