Discrete Directivity Encoding for Interpolation-Free 6 DoF Rendering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional discrete directivity representations for sound sources are sub-optimal for 6 Degrees-of-Freedom (6 DoF) rendering, requiring interpolation and resulting in large bitstream sizes due to redundancy and irrelevance in directivity data.
Innovation Solution
A method for optimizing directivity representation by distributing unit vectors non-uniformly on a 3D sphere using a predetermined arrangement algorithm, allowing for efficient decoding and reduced bitrates without interpolation, based on perceptual directivity sensitivity thresholds.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If conventional discrete directivity representations are used, then directivity information can be transmitted, but the bitstream size becomes large due to redundancy and irrelevance
Solution Approach 1:
The patent extracts only the essential directivity information by representing the radiation pattern through a reduced set of parameters (spherical harmonics coefficients, discrete vector data, or parametric models) rather than transmitting complete measured data. This extraction removes redundant and imperceptible components, achieving compact representation while preserving perceptually relevant directivity characteristics.
Solution Approach 2:
The patent transforms directivity representation from raw measured data to optimized parameter sets (spherical harmonics coefficients, discrete vectors with gains, or parametric model parameters). This parameter transformation reduces data dimensionality while maintaining the ability to reconstruct the radiation pattern, thereby reducing bitstream size without significant loss of perceptual quality.
2Measurement precision
If conventional discrete directivity representations are used, then directivity data can be transmitted, but computational complexity increases due to interpolation requirements
Solution Approach 1:
The patent performs preliminary organization of directivity data into structured formats (spherical harmonics decomposition, uniformly distributed discrete vectors, or parametric models) during encoding. This preliminary structuring enables efficient retrieval and rendering without requiring complex interpolation operations during real-time playback, thereby reducing computational complexity while maintaining accuracy.
Solution Approach 2:
The patent creates simplified copies of the radiation pattern using mathematical models (spherical harmonics, discrete vectors, or parametric representations) that capture the essential directivity characteristics. These model-based copies can be evaluated directly without interpolation, reducing computational complexity compared to working with complete measured data sets.
3Measurement precision
If non-uniformly distributed unit vectors are used, then representation accuracy improves, but encoding complexity increases
Solution Approach 1:
The patent employs adaptive strategies where the number and distribution of discrete vectors are dynamically adjusted based on the complexity of the radiation pattern and perceptual requirements. For simple patterns, fewer vectors are used; for complex patterns, more vectors are allocated. This dynamic adaptation optimizes the balance between representation accuracy and encoding complexity.
Solution Approach 2:
The patent transforms the directivity representation into parameterized forms (spherical harmonics coefficients, discrete vectors with associated gains, or parametric model parameters) that can efficiently represent non-uniform distributions. These parameter changes enable accurate representation of angular variations without requiring explicit storage of all individual vector directions, thereby managing encoding complexity.
Data Source
AI summary
The present disclosure relates to a method of processing audio content including directivity information for at least one sound source, the directivity information comprising a first set of first directivity unit vectors representing directivity directions and associated first directivity gains. The disclosure further relates to corresponding methods of encoding and decoding audio content including directivity information for at least one sound source.