Volumetric Dataset Compression via Uniform Mesh Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional systems face inefficiencies in representing and storing volumetric three-dimensional data, particularly in generating and processing datasets for electronic devices, as they require large storage sizes and struggle with temporal coherence, leading to inconsistent mesh representations that hinder dynamic effects and efficient data compression.
Innovation Solution
The system processes volumetric data to generate a reduced-size, temporally coherent uniform topology and UV atlas, utilizing optical flow points and image data to track and optimize mesh topology, resulting in a significantly smaller storage size, approximately one-tenth of the original, while enabling efficient application of graphical and animation effects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If conventional volumetric data representation methods are used, then data completeness is maintained, but storage size becomes excessively large
Solution Approach 1:
The patent extracts and stores only the essential geometric features (vertices, edges, faces) of the volumetric mesh rather than complete voxel data. By taking out only the necessary topological information and using UV atlas mappings to represent surface properties, the system achieves significant compression while maintaining visual fidelity and enabling dynamic effects.
Solution Approach 2:
Instead of storing complete volumetric data and compressing it, the patent inverts the approach by constructing a simplified mesh representation from volumetric samples and storing only the essential geometric parameters. This inverted workflow achieves better compression ratios while preserving the necessary information for rendering and animation.
2Stability of the object's composition
If mesh topology changes dynamically to represent temporal variations, then temporal coherence is improved, but data consistency deteriorates
Solution Approach 1:
The patent performs preliminary mesh generation and UV atlas creation from a reference frame before processing temporal variations. By establishing the mesh topology and UV mapping in advance, the system maintains consistent data structures across all frames while capturing temporal changes through vertex displacement and texture animation, ensuring both temporal coherence and data consistency.
Solution Approach 2:
The patent implements dynamic mesh deformation techniques where the base mesh topology remains fixed but vertex positions and UV coordinates are animated to represent temporal variations. This dynamic approach allows the mesh to adapt to different poses and expressions while maintaining topological consistency, resolving the contradiction between temporal coherence and data consistency.
3Manufacturing precision
If high-resolution volumetric data is stored, then visual quality is maintained, but processing efficiency deteriorates
Solution Approach 1:
The patent segments the volumetric data into discrete mesh elements (vertices, edges, faces) and further segments the surface representation into UV atlas regions. This segmentation allows selective processing of different mesh components and enables parallel processing during rendering and animation, significantly improving processing efficiency while maintaining visual quality through preserved geometric detail.
Solution Approach 2:
The patent creates a simplified mesh copy from the volumetric data that serves as a lightweight representation for processing. This mesh copy contains only the essential geometric information needed for rendering and animation, allowing efficient processing while the original volumetric data can be referenced when needed for high-quality visualization, thus resolving the efficiency-quality tradeoff.
Data Source
AI summary
Device, system, and method of generating a reduced-size volumetric dataset. A method includes receiving a plurality of three-dimensional volumetric datasets that correspond to a particular object; and generating, from that plurality of three-dimensional volumetric datasets, a single uniform mesh dataset that corresponds to that particular object. The size of that single uniform mesh dataset is less than ¼ of the aggregate size of the plurality of three-dimensional volumetric datasets. The resulting uniform mesh is temporally coherent, and can be used for animating that object, as well as for introducing modifications to that object or to clothing or garments worn by that object.