Volumetric Video Tiling for 3DoF+ Rendering With Lower Bandwidth

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing technologies lack a stream and associated methods that can carry data representative of a volumetric scene, which can be encoded at once and decoded either as 3DoF video or as volumetric video (3DoF+ or 6DoF), and the amount of data required for rendering on end-user devices increases bandwidth needs significantly.

Innovation Solution

A method and device for encoding and decoding volumetric video data that allows for 3DoF and 3DoF+ rendering by partitioning frames into tiles, associating texture and depth information, and using 2D parametrization to represent 3D scene parts, enabling flexible rendering on immersive devices.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If volumetric video data is transmitted using conventional methods, then compatibility with existing video processing systems is maintained, but the data volume becomes excessively large and transmission time becomes unacceptably long

Engineering Contradiction:

Improvedata volumeVSAvoidtransmission time

Core Design Contradiction:

Quantity of substanceVSLoss of time

Solution Approach 1:

The volumetric video data is segmented into multiple view images corresponding to different viewing angles. Each view image is processed and transmitted separately, allowing for reduced data volume per stream while maintaining the complete volumetric video information when all segments are reassembled.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent transforms the traditional single-view video representation into a multi-dimensional volumetric representation by adding the viewing angle dimension. This allows the same visual information to be represented more efficiently through structured angular sampling rather than transmitting all possible pixel data.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

2Loss of information

If volumetric video data is transmitted using conventional methods, then complete video information is preserved, but the data volume becomes excessively large

Engineering Contradiction:

Improvevideo information completenessVSAvoiddata volume

Core Design Contradiction:

Loss of informationVSQuantity of substance

Solution Approach 1:

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Different viewing angles are sampled with appropriate resolution based on their importance and visual characteristics. This allows optimization of data quality where needed while reducing data volume in less critical angular regions, maintaining overall information completeness without uniform high-quality transmission across all views.

Inventive Principle:

Principle #3Local quality

3Adaptability or versatility

If conventional video processing systems are used, then existing infrastructure is utilized, but the systems cannot properly process volumetric video data

Engineering Contradiction:

Improvesystem compatibilityVSAvoidvideo processing capability

Core Design Contradiction:

Adaptability or versatilityVSProductivity

Solution Approach 1:

The patent designs a video processing system that can handle both conventional single-view videos and volumetric multi-view videos using the same fundamental architecture. The system universally processes video data by accepting different input formats (single view or multiple angular views) and producing appropriate output, thereby maintaining compatibility with existing infrastructure while enabling new volumetric video processing capabilities.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentEP3777222B1Method, apparatus and stream for volumetric video format

Publication Date: 2026.05.06 INTERDIGITAL MADISON PATENT HLDG

Figure 1~2
Figure 3A~4
Figure 5~6

AI summary

Encoding/decoding data representative of a 3D scene can involve first data representing texture encoded into a first track and arranged in first tiles of a first frame, a part of the 3D scene being associated with each first tile; second data representing depth encoded into one or more second tracks and arranged in second tiles of a second frame greater in number than the number of first tiles, second tiles being allocated to each first tile and patches are arranged in the set of second tiles corresponding to a 2D parametrization of a group of 3D points in the part of the 3D scene associated with each first tile and comprising second data representing depth associated with the 3D points of the group; and instructions encoded into a third track to extract part of the first and second data.