Parallelogram Grid Prediction for Efficient 3D Mesh Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing 3D mesh encoding and decoding technologies are inefficient in utilizing redundancies and correlations within 3D representations, leading to high data intensity and resource consumption, particularly in immersive applications.
Innovation Solution
The use of a predictor from a reference pool with candidate positions on a parallelogram grid to adaptively determine reference positions for encoding and decoding 3D meshes, allowing for efficient encoding and decoding without additional signaling beyond an index.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If traditional 3D mesh encoding methods are used, then encoding and decoding can be performed, but data intensity and resource consumption are high
Solution Approach 1:
The patent applies preliminary action by pre-establishing a reference pool containing multiple candidate reference positions (including original position, left neighbor, right neighbor, and diagonal positions) before encoding. This allows the encoder to select the optimal reference position for prediction without needing complex real-time calculations, thereby reducing data intensity while maintaining encoding efficiency
Solution Approach 2:
The patent changes the parameter approach by transitioning from storing complete 3D mesh data to storing only the residual information after prediction. By using parallelogram grid-based prediction to estimate the current position from reference positions, the system encodes only the difference (residual), significantly reducing the quantity of data while preserving encoding accuracy
2Productivity
If adaptive reference position selection is implemented, then encoding efficiency improves, but device complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the reference selection process into discrete, manageable components: a fixed set of candidate reference positions (original, left, right, diagonal), a parallelogram grid structure for organization, and a simple index-based selection mechanism. This segmented approach reduces the complexity of adaptive reference selection while maintaining high encoding efficiency through structured candidate generation
Solution Approach 2:
The patent uses copying by creating a reference pool that copies and stores multiple potential reference positions (original position, left neighbor, right neighbor, diagonal positions) in advance. This pre-computed reference pool allows the encoder to simply select from copies rather than perform complex real-time calculations, reducing device complexity while maintaining adaptive selection benefits
3Measurement precision
If parallelogram grid prediction is used, then prediction accuracy improves, but computational requirements increase
Solution Approach 1:
The patent applies partial action by using a selective approach where the system evaluates only the necessary reference positions (original, left, right, and diagonal neighbors) within the parallelogram grid structure rather than computing all possible combinations. This partial evaluation of the reference pool achieves sufficient prediction accuracy while significantly reducing computational energy consumption compared to exhaustive methods
Data Source
AI summary
This disclosure relates generally to encoding and decoding of 3-dimensional (3D) mesh and is particularly directed to predicting a 3D mesh position using a predictor from a reference pool having candidate reference positions on a parallelogram grid. For example, when encoding a current position, a set of prior encoded positions may be used to derive a set of reference positions that are distributed as grid points in in a parallelogram grid space that encompasses a parallelogram prediction position. The set of reference positions may be adaptively determined. The reference set may be derivable by both the encoder and the decoder, thereby requiring no signaling other than an index for an optimal predictor selected from the set of reference positions for encoding the current position.