Video Partition Signaling Using Scan-Order Tile and Brick Assignment

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The complex syntax structure for signaling tile and brick partitioning in video coding standards like VVC requires an inefficient bitrate, which needs optimization.

Innovation Solution

A method and apparatus for determining and assigning video units to partitions in a predefined scan order, including units such as tile columns, rows, and brick rows, while marking unassigned units to last partitions, optimizing the partitioning process.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If a complex syntax structure is used for signaling tile and brick partitioning options, then more partitioning configurations can be supported, but the bitrate required for signaling increases

Engineering Contradiction:
Improvepartitioning configuration optionsVSAvoidbitrate
Core Design Contradiction:
Adaptability or versatilityVSQuantity of substance

Solution Approach 1:

The partitioning signaling is segmented into multiple stages: first signaling the number of partitions, then signaling sizes for explicitly sized partitions, and finally assigning remaining units to partitions in a predefined scan order. This segmentation allows the syntax structure to support diverse partitioning configurations while transmitting information in an optimized sequence that reduces overall bitrate requirements.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary actions by first indicating the number of partitions and their sizes before assigning specific units to partitions. This preliminary signaling of partition structure allows the decoder to prepare the partitioning framework in advance, enabling more efficient subsequent processing and reducing the bitrate needed for detailed partition assignment information.

Inventive Principle:
Principle #10Preliminary action

2Adaptability or versatility

If multiple partitioning options are signaled explicitly, then coding flexibility is improved, but the syntax complexity increases

Engineering Contradiction:
Improvepartitioning flexibilityVSAvoidsyntax structure
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The syntax structure is made dynamic by allowing the number of partitions and their sizes to be signaled adaptively based on the specific partitioning scenario. Rather than a fixed complex structure for all cases, the syntax dynamically adjusts to signal only the necessary partitioning information, maintaining flexibility while reducing structural complexity.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes parameters such as the number of partitions and their sizes based on the partitioning requirements. By signaling these parameters adaptively rather than using a fixed complex structure, the syntax maintains coding flexibility while avoiding unnecessary structural complexity. The predefined scan order for assigning units to partitions further simplifies the syntax by eliminating the need for complex assignment signaling.

Inventive Principle:
Principle #35Parameter changes

3Manufacturing precision

If all partition assignments are signaled explicitly, then partitioning precision is improved, but the signaling bitrate increases

Engineering Contradiction:
Improvepartition assignment precisionVSAvoidsignaling bitrate
Core Design Contradiction:
Manufacturing precisionVSQuantity of substance

Solution Approach 1:

The patent extracts and separates the partitioning information into two parts: explicitly signaled partition sizes and implicitly assigned unit allocations. By taking out the explicit size signaling and handling the remaining unit assignments through a predefined scan order, the patent achieves precise partition assignment while reducing the bitrate required for signaling, as the scan order implicitly defines the assignment without requiring additional explicit signals.

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentUS12483702B2Apparatus, a method and a computer program for video coding and decoding
Publication Date: 2025.11.25 NOKIA TECHNOLOGIES OY
  • US12483702B2 patent drawing
  • US12483702B2 patent drawing
  • US12483702B2 patent drawing

AI summary

A method comprising: determining a number of units that are to be assigned to partitions and are initialized as unassigned; indicating or inferring a number of explicitly sized partitions to be assigned; indicating sizes for the explicitly sized partitions and accordingly marking unassigned units as assigned to partitions in a pre-defined scan order; indicating a count of units; repetitively assigning the count of units to partitions and accordingly marking unassigned units as assigned in the pre-defined scan order until the number of unassigned units is less than the count of units; and assigning, if the number of unassigned units is greater than 0, the unassigned units to a last partition.