Geometry Encoding with ML-Based Compression Parameter Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Selecting optimal compression techniques and parameters for geometric data, such as mesh data, becomes increasingly complex due to the variety and complexity of compression models, making it difficult for users to achieve optimal compression results.
Innovation Solution
A system and method that uses machine learning to select compression techniques and parameters by generating a signature based on geometric data properties, training a classifier, and enumerating options to determine the best encoding and decoding settings, allowing for automatic selection of optimal compression parameters without requiring extensive user knowledge.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If the number and variety of compression models increase to improve compression performance, then compression quality is improved, but the complexity of selecting optimal models and parameters increases
Solution Approach 1:
The system automatically selects compression models and parameters by analyzing geometric data properties (number of vertices, edges, triangles, connected components, boundary edges, angle histograms, vertex valences) and using machine learning classifiers to determine optimal encoding options without requiring user intervention or manual configuration
Solution Approach 2:
The system changes the approach from manual parameter selection to automated parameter determination by using machine learning models that take geometric data properties as input and output optimal compression parameters, effectively transforming the selection process from user-driven to system-driven
2Manufacturing precision
If the number of parameters associated with compression models increases to improve compression performance, then compression quality is improved, but the difficulty of selecting desirable compression models increases
Solution Approach 1:
The system performs self-service by automatically analyzing geometric data properties and selecting optimal compression parameters without requiring user knowledge or manual configuration, making the system easy to use while maintaining high compression quality
Solution Approach 2:
The machine learning classifier acts as an intermediary between the geometric data properties and the compression parameters, translating data characteristics into optimal encoding options automatically, thereby eliminating the need for users to directly manage complex parameter selections
Data Source
AI summary
A method includes receiving geometric data to be encoded, generating a signature for the geometric data based on the at least one property associated with the geometric data, enumerating a set of first options, enumerating a set of second options, encoding the geometric data using the enumerated first option and the enumerated second option, decoding the encoded geometric data, selecting one of the enumerated second options based on a cost function, and training a classifier based on the signature, the enumerated first option and the selected second option.