Base Mesh Entropy Coding With Fixed-Point Vertex Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current dynamic mesh coding standards, such as V-DMC TMM 7.0, face inefficiencies in position and material property prediction during base-mesh entropy encoding, leading to high computational requirements and inconsistent results across platforms.
Innovation Solution
Implement a processor-based apparatus and method for dynamic mesh coding that utilizes neighboring vertex coordinates and weights to predict geometric and texture coordinates, reducing the need for 32-bit or 64-bit floating point numbers and ensuring consistent results across platforms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If 32-bit or 64-bit floating point numbers are used for position and material property prediction, then prediction precision is improved, but computational complexity and processing time increase
Solution Approach 1:
The patent changes the parameter representation from 32-bit or 64-bit floating point numbers to fixed-point arithmetic with 7 or 8 bits. This parameter transformation maintains sufficient prediction precision for mesh vertex positions and material properties while dramatically reducing computational complexity and processing requirements.
Solution Approach 2:
The patent employs simplified fixed-point prediction values that are computationally inexpensive to calculate and use. These simplified prediction values replace complex floating-point calculations, providing adequate precision for entropy coding applications without the high computational cost of floating-point arithmetic.
2Adaptability or versatility
If platform-specific floating point implementations are used, then computational flexibility is improved, but result consistency across platforms deteriorates
Solution Approach 1:
The patent transforms the computational approach from platform-dependent floating-point arithmetic to platform-independent fixed-point arithmetic. By changing the numerical representation to fixed-point with predetermined bit allocations (7 or 8 bits), the patent ensures that position and material property predictions produce consistent results across different computing platforms while maintaining adequate precision.
3Measurement precision
If detailed position prediction using multiple neighboring vertices is implemented, then prediction accuracy is improved, but processing time increases
Solution Approach 1:
The patent uses fixed-point arithmetic with predetermined bit lengths (7 or 8 bits) for position prediction calculations involving multiple neighboring vertices. This parameter constraint enables efficient computation of accurate predictors by limiting the precision to what is actually needed for entropy coding, avoiding unnecessary computational overhead while maintaining sufficient prediction accuracy.
Data Source
AI summary
An apparatus directed to improvements to motion coding for vertices in an inter-coded base mesh frame is provided. The apparatus receives a compressed bitstream comprising a base mesh sub-bitstream and a syntax element. The apparatus decodes the syntax element to generate a prediction error. The apparatus decodes the base mesh sub-bitstream to generate a base mesh frame. The apparatus determines neighboring vertices of a vertex based on the base mesh frame. The apparatus determines a first predictor based on the neighbors of the vertex. The apparatus determines the vertex based on the first predictor and the prediction error.