glTF Polygon Encoding for Mixed-Topology Scene Descriptions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The current MPEG-I Scene Description (SD) format is limited in encoding 3D meshes, allowing only triangles and lacks flexibility in encoding polygons of varying sizes and mixed topologies, requiring new meshes and nodes for additional topologies.
Innovation Solution
A new glTF extension, 'MPEG_accessor_tuples', is introduced to store a list of tuples for vertex indices, and 'MPEG_primitive_topologies' to provide additional topologies, enabling encoding of polygons of any size and mixed topologies without replacing the default triangle-based encoding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the current MPEG-I Scene Description format is used to encode 3D meshes, then the encoding process is simple and follows a standard format, but the format is limited to triangles only and lacks flexibility for encoding polygons of varying sizes and mixed topologies
Solution Approach 1:
The patent segments the polygon encoding into two distinct data structures: a first data structure for standard accessor information and a second data structure for MPEG_accessor_tuples extension properties. This segmentation allows the system to maintain both simple triangle encoding and flexible polygon encoding without mixing their respective data requirements, thereby improving adaptability while managing complexity through structured separation.
Solution Approach 2:
The patent creates a universal tuples data structure that can handle multiple topology types (triangles, quads, n-gons) within a single encoding framework. The MPEG_accessor_tuples extension properties enable the same data structure to serve both traditional triangle meshes and advanced polygonal meshes, providing multi-functionality that enhances encoding flexibility without requiring separate specialized structures for each topology type.
2Adaptability or versatility
If additional topologies are encoded using the current format, then new meshes and nodes must be created for each topology type, but this increases the complexity of scene description and requires additional knowledge to interpret
Solution Approach 1:
The patent introduces MPEG_accessor_tuples extension properties as an intermediary layer between the standard accessor information and the actual topology data. This intermediary structure provides a standardized interface that clearly defines how to interpret different topology types, eliminating the need for additional interpretation knowledge while supporting multiple topologies within a single mesh primitive.
Solution Approach 2:
The patent performs preliminary organization of topology data by storing edges and indices in predetermined locations within the tuples data structure before rendering or processing. The edges property and indices property are pre-arranged in specific formats, allowing downstream processes to directly interpret the data without requiring additional knowledge about the intended topology, thus preventing information loss.
3Adaptability or versatility
If the glTF format is extended to support polygons of any size, then the scene description capabilities are enhanced, but the data structure complexity increases due to multiple extension properties
Solution Approach 1:
The patent implements a nested data structure where the MPEG_accessor_tuples extension properties are embedded within the existing accessor information framework. The first data structure (standard accessor) contains a reference to or is complemented by the second data structure (extension properties), creating a nested organization that allows polygon encoding capabilities to be added without fundamentally restructuring the entire glTF format, thereby managing data structure complexity.
Data Source
Figure 1A
Figure 1B
Figure 1C
AI summary
Some embodiments of a method may include obtaining information corresponding to a mesh associated with an object in a 3D scene representation; parsing accessor information, wherein the information corresponding to the object comprises the accessor information; storing a first portion of the accessor information in a first data structure; parsing the accessor information for MPEG_accessor_tuples extension properties; storing a second portion of the accessor information in a second data structure, wherein the second portion of accessor information corresponds to the MPEG_accessor_tuples extension properties; building a tuples data structure using the first and second data structures; and outputting the tuples data structure.