Layered Volumetric Video Bitstream Signaling and Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current volumetric video coding technologies, such as V3C, fail to effectively identify and manage layered video components within bitstreams, leading to inefficiencies in decoding and rendering, particularly when redundant source representations are involved, as they lack signaling information about layer relationships and constraints.

Innovation Solution

The method involves encoding volumetric video components using a layered video encoder, maintaining information about layer relationships, and encapsulating these components into V3C units with signaling information that allows for decoding and rendering of complete or partial bitstreams, ubiquitously, by providing signaling elements that indicate layer relationships and constraints.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If V3C video components are encoded by a layered video codec and then split into V3C unit payloads, then the components can be stored in independent payloads for flexible processing, but it becomes impossible to reconstruct the original layered video coded bitstream and lose the benefits of layered video coding

Engineering Contradiction:
Improveflexibility in processing and storing video componentsVSAvoidloss of layered video coding benefits and inability to reconstruct original bitstream
Core Design Contradiction:
Adaptability or versatilityVSLoss of information

Solution Approach 1:

The patent segments the layered video bitstream into independent V3C unit payloads, each containing specific video components (texture, geometry, occupancy) at different layers. This segmentation allows flexible processing and storage while maintaining the ability to reconstruct the original bitstream through proper reassembly of the segmented units.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediary data structure that preserves the layered video coding information during the transition from layered video codec output to independent V3C unit payloads. This intermediary structure maintains the relationships between layers and components, enabling reconstruction of the original bitstream while allowing independent processing of each payload.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Ease of operation

If V3C bitstream stores video components within independent V3C unit payloads, then processing and transmission become more flexible, but the decoding and rendering efficiency decreases due to lack of layer relationship information

Engineering Contradiction:
Improveflexibility in processing and transmissionVSAvoiddecoding and rendering efficiency
Core Design Contradiction:
Ease of operationVSProductivity

Solution Approach 1:

The patent performs preliminary organization of video components into properly structured V3C unit payloads that include all necessary layer relationship information. By pre-organizing the data with embedded metadata about layer dependencies and component relationships, the decoding and rendering processes can proceed efficiently without needing to perform complex analysis during playback.

Inventive Principle:
Principle #10Preliminary action

3Adaptability or versatility

If multiple redundant source representations are captured from different viewing angles, then the volumetric video content becomes more comprehensive, but the data redundancy increases significantly

Engineering Contradiction:
Improvecompleteness of volumetric video contentVSAvoidamount of data to be processed
Core Design Contradiction:
Adaptability or versatilityVSQuantity of substance

Solution Approach 1:

The patent merges multiple redundant source representations from different viewing angles into a unified layered video structure. By combining the redundant information into a single organized bitstream with proper layer relationships, the system maintains comprehensive volumetric content while reducing the effective data quantity through efficient representation and elimination of true redundancy.

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentEP4145832A1An apparatus, a method and a computer program for volumetric video
Publication Date: 2023.03.08 NOKIA TECHNOLOGIES OY
  • EP4145832A1 patent drawingFigure 1a
  • EP4145832A1 patent drawingFigure 1b
  • EP4145832A1 patent drawingFigure 2a

AI summary

There is disclosed a method comprising: obtaining bitstreams of video components of a layered volumetric video; analyzing the video components to determine a relationship between the video components and corresponding layers; providing signaling elements to indicate if the volumetric video comprises video components at different layers; providing in the signaling elements information of the relationship between the video components and corresponding layers; encapsulating the layered video coded bitstreams to visual volumetric video-based coding units; and constructing a bitstream according to specified constraints. There is also disclosed a method comprising: receiving a bitstream comprising video components of a layered volumetric video; demultiplexing video components of the volumetric video; receiving signaling elements; examining the signaling elements to determine if the volumetric video comprises video components at different layers; extracting from the one or more signaling elements information of relationship between the video components and corresponding layers; and re-multiplexing the layered video components into one layered video bitstream for video decoding.