Point Cloud Rendering Redundant Pixel Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current projection-based methods for compressing point cloud data introduce 30%-60% redundant data, leading to reduced performance in real-time rendering of dynamic point clouds on mobile devices due to inefficient GPU rendering pipelines.
Innovation Solution
A method that filters out redundant pixels from 2D image frames based on occupancy metadata, reconstructs 3D points, and generates texture images to optimize the rendering process by discarding redundant data and projecting non-redundant points to a screen space domain for efficient rendering.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If projection-based methods are used to compress point cloud data, then data compression is achieved, but 30%-60% redundant data is introduced leading to reduced rendering performance
Solution Approach 1:
The patent extracts and removes redundant pixels from the point cloud data before rendering. By identifying and eliminating duplicate pixel information that arises from projection-based compression, the system processes only unique data points, thereby maintaining high rendering performance while achieving data compression.
Solution Approach 2:
The patent performs preliminary filtering of redundant pixels before the rendering process. By pre-processing the point cloud data to remove duplicates and redundancies ahead of time, the system prepares optimized data structures that enable fast real-time rendering without the performance penalty of processing redundant information.
2Reliability
If GPU rendering pipeline processes all pixels including redundant ones, then complete data processing is achieved, but processing time increases reducing real-time rendering capability
Solution Approach 1:
The patent extracts redundant pixels from the input data before they enter the GPU rendering pipeline. By removing duplicate pixel information in advance, the system ensures that only necessary data is processed, maintaining processing completeness for valid data while significantly reducing the total number of pixels that require GPU processing time.
Solution Approach 2:
The patent discards redundant pixel data that would otherwise waste processing time. By identifying and discarding duplicate pixels through occupancy metadata analysis, the system recovers valuable processing time while maintaining the integrity and completeness of processing all unique data points.
3Loss of information
If redundant pixels are processed in GPU rendering pipeline, then no data is lost, but rendering efficiency decreases making 60 fps real-time rendering difficult on mobile devices
Solution Approach 1:
The patent extracts and removes redundant pixels from the point cloud data before rendering. By eliminating duplicate pixel information that provides no additional visual value, the system prevents data loss of meaningful information while dramatically improving rendering efficiency to enable real-time 60 fps performance on mobile devices.
Solution Approach 2:
The patent changes the parameter of pixel redundancy by using occupancy metadata to identify and remove duplicate pixels. This parameter transformation converts a high-redundancy data structure into a low-redundancy structure, maintaining visual information integrity while boosting rendering efficiency for real-time mobile applications.
Data Source
AI summary
A method and apparatus for rendering point cloud-based volumetric data are provided. The method includes receiving, at a computing device, a plurality of two-dimensional (2D) frames associated with a three-dimensional (3D) point cloud, and projection metadata comprising details of projection of the 3D point cloud into a 2D image plane; reconstructing, by the computing device, a plurality of 3D points associated with the 3D point cloud, based on the received 2D frames and the projection metadata; and generating texture images based on the plurality of 3D points; and rendering at least one quad image based on the texture images.