Basemesh HRD Buffering for Independent Submesh Decoding

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The existing decoding processes for basemesh bitstreams in the new international dynamic mesh compression standard (ISO/IEC 23090-29) face challenges due to the lack of a unified buffering management system, which complicates the application of HRD models and requires separate processing for intra and inter encoded meshes, leading to inefficiencies.

Innovation Solution

A Hypothetical Reference Decoding (HRD) operation is introduced to manage basemesh bitstreams with a single buffering system, utilizing a Coded BaseMesh Buffer (CBMB) and Decoded BaseMesh Buffer (DBMB) to enable independent decoding of submeshes, allowing for efficient processing without internal buffering and aligning with video decoder architectures.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If separate processing for intra and inter encoded meshes is used, then decoding accuracy is maintained, but device complexity and processing overhead increase

Engineering Contradiction:
Improvedecoding accuracyVSAvoidbuffering management complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent merges the separate buffering management systems for intra and inter encoded meshes into a unified HRD model. The Coded BaseMesh Buffer (CBMB) and Decoded BaseMesh Buffer (DBMB) are managed through a single HRD framework that handles both encoding types, eliminating the need for separate buffer management while maintaining decoding accuracy through consistent reference mesh provisioning.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The HRD model is designed to serve multiple functions simultaneously: it manages both intra and inter encoded meshes, provides reference meshes for both encoding types, and handles timing and synchronization for diverse mesh data. This universal buffering system replaces multiple specialized buffers, reducing overall system complexity while maintaining all required decoding capabilities.

Inventive Principle:
Principle #6Universality (Multi-functionality)

2Reliability

If internal buffering is utilized for processing NAL units, then decoding completeness is ensured, but productivity and processing efficiency decrease

Engineering Contradiction:
Improvedecoding completenessVSAvoidprocessing efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent extracts the buffering function from the internal processing path and places it in the external HRD model. The CBMB and DBMB are defined as external entities that the decoder can utilize without implementing complex internal buffering logic. This extraction maintains decoding completeness by ensuring all necessary reference meshes are available while simplifying the decoder's processing path and improving efficiency.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The HRD model acts as an intermediary between the bitstream and the decoder, managing all buffering operations externally. The CBMB and DBMB serve as intermediary structures that the decoder accesses through standardized interfaces, eliminating the need for the decoder to implement its own internal buffering mechanisms and thereby improving processing efficiency while maintaining completeness.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Device complexity

If the video HRD model is applied to basemesh sub-bitstream, then buffering management is simplified, but adaptability to mesh-specific decoding requirements is lost

Engineering Contradiction:
Improvebuffering management simplicityVSAvoidmesh decoding adaptability
Core Design Contradiction:
Device complexityVSAdaptability or versatility

Solution Approach 1:

The patent adapts the video HRD model by introducing mesh-specific parameters and structures while maintaining the overall simplified buffering framework. The CBMB and DBMB are configured with mesh-specific attributes such as vertex information, connectivity data, and mesh-specific timing parameters. This allows the system to maintain the simplicity of the HRD model while being fully adaptable to mesh decoding requirements through localized modifications.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent modifies the HRD model parameters to accommodate mesh-specific requirements. The buffering structures include mesh-specific parameters such as vertex coordinate information, connectivity data, and mesh timing information. These parameter changes enable the simplified HRD framework to handle diverse mesh decoding scenarios, from static to dynamic meshes, while maintaining buffering management simplicity.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS20260017829A1Hypothetical reference decoding operation for basemesh bitstreams
Publication Date: 2026.01.15 SONY GROUP CORP
  • US20260017829A1 patent drawing
  • US20260017829A1 patent drawing
  • US20260017829A1 patent drawing

AI summary

A Hypothetical Reference Decoding (HRD) operation for basemesh bitstreams is described herein. An HRD model is able to handle and address the peculiarities associated with the decoding process, and provisioning of appropriate HRD related syntax structures and tables in the Video Usability information (VUI) as well as buffering period, basemesh timing, and decoding unit information Supplemental Enhancement Information (SEI) messages.