Dynamic Mesh Coding Using Unified Texture and Displacement Streams

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing dynamic mesh coding systems require separate encoders/decoders for texture and displacement data, and displacement data is often coded in formats that demand excessive encoding/decoding resources.

Innovation Solution

Combine texture and displacement data into a single bitstream, using different coding methods for each, and apply techniques like conversion to 4:2:0 format, padding, and smoothing to optimize encoding/decoding efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If texture data and displacement data are coded separately using different coding methods, then coding flexibility and quality are improved, but device complexity and resource usage increase due to requiring multiple encoders/decoders

Engineering Contradiction:

Improvecoding qualityVSAvoidnumber of encoders/decoders

Core Design Contradiction:

Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent combines texture data and displacement data into a single bitstream, allowing them to be transmitted and stored together while maintaining the ability to apply different coding methods to each data type. This merging approach reduces the number of separate encoders/decoders needed while preserving coding flexibility.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

Within the unified bitstream, the patent segments the coding approach by applying different coding methods to different data types - specifically using lossless coding for displacement data and configurable coding methods for texture data. This segmentation allows optimized handling of each data type's unique requirements.

Inventive Principle:

Principle #1Segmentation

2Measurement precision

If displacement data is coded in high-precision formats, then measurement precision is improved, but resource usage and encoding/decoding complexity increase

Engineering Contradiction:

Improvedisplacement data precisionVSAvoidencoding/decoding efficiency

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The patent changes the coding parameters for displacement data by applying lossless coding methods that preserve high precision while optimizing the encoding process. This allows maintaining measurement precision without the excessive resource usage associated with traditional high-precision formatting approaches.

Inventive Principle:

Principle #35Parameter changes

3Device complexity

If a single bitstream is used for both texture and displacement data, then device complexity is reduced, but coding flexibility may be compromised

Engineering Contradiction:

Improvenumber of encoders/decodersVSAvoidcoding method flexibility

Core Design Contradiction:

Device complexityVSAdaptability or versatility

Solution Approach 1:

The patent introduces dynamic adaptability within the single bitstream framework by allowing different coding methods to be applied to different portions of data. The encoder can dynamically select appropriate coding methods for texture and displacement data, and the decoder can adaptively process different data types with their respective optimal methods, maintaining flexibility within the unified structure.

Inventive Principle:

Principle #15Dynamics

Data Source

PatentUS20250343931A1Jointly coding of texture and displacement data in dynamic mesh coding

Publication Date: 2025.11.06 BYTEDANCE INC

AI summary

A mechanism for processing video data is disclosed. The mechanism includes determining that the texture data and the displacement data are included in a single bitstream and use different coding methods. A conversion is performed between the visual media data and the single bitstream based on the different coding methods of the texture data and the displacement data.