Dynamic Mesh Compression via 2D UV Atlas Sampling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing mesh compression standards fail to efficiently compress dynamic 3D meshes with time-varying attribute maps and connectivity information, particularly under real-time constraints, and do not support volumetric acquisition techniques.
Innovation Solution
A sampling-based dynamic mesh compression method that determines sample positions, occupancy status, and generates a sample-based occupancy map, which is then compressed using a video codec into single-channel or multiple-channel images, enabling efficient storage and transmission of dynamic meshes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If existing mesh compression standards (IC, MESHGRID, FAMC) are used, then constant connectivity dynamic meshes can be compressed, but time varying attribute maps and connectivity information cannot be handled
Solution Approach 1:
The patent segments the mesh compression problem into multiple independent components: occupancy map generation, geometry map generation, and attribute map generation. Each component handles a specific aspect of the mesh data separately, allowing the system to process time-varying attribute maps and connectivity information without compromising compression accuracy.
Solution Approach 2:
The patent transforms 3D mesh data into 2D occupancy maps and attribute maps through UV coordinate transformation. This dimensional reduction enables efficient compression using video codecs while preserving the essential information about time-varying connectivity and attributes, resolving the contradiction between adaptability and reliability.
2Speed
If volumetric acquisition techniques are used to generate dynamic meshes, then real time performance can be achieved, but constant connectivity cannot be maintained
Solution Approach 1:
The patent explicitly designs the compression system to handle dynamic connectivity changes. The occupancy map generation process dynamically determines which vertices are active in each frame, allowing the system to accommodate real-time volumetric acquisition results even when connectivity varies over time.
Solution Approach 2:
The patent changes the representation parameters from fixed connectivity indices to occupancy-based masks. This parameter transformation allows the system to maintain stable compression performance while adapting to dynamic connectivity changes introduced by volumetric acquisition techniques.
3Productivity
If dynamic mesh data is compressed using traditional methods, then data size can be reduced, but compression efficiency for time varying content is insufficient
Solution Approach 1:
The patent creates a simplified representation (occupancy map) that copies only the essential information about vertex occupancy and connectivity changes. This selective copying approach reduces data size significantly while maintaining compression efficiency, as the occupancy maps can be processed using efficient video coding techniques.
Solution Approach 2:
The patent extracts the time-varying connectivity and attribute information from the full mesh data and separates it into independent occupancy maps and attribute maps. This extraction allows each component to be compressed independently using optimized methods, improving overall compression efficiency while reducing the quantity of data that needs to be stored and transmitted.
Data Source
AI summary
Method, apparatus, and system for sampling-based dynamic mesh compression are provided. The process may include determining one or more sample positions associated with an input mesh based on one or more sampling rates, and an occupancy status associated respectively with each of the one or more sample positions indicating whether each of the one or more sample positions is within boundaries of one or more polygons defined by the input mesh is determined. The process may include generating a sample-based occupancy map based on the occupancy status associated respectively with each of the one or more sample positions.