Video Slice Partitioning for Efficient High-Definition Decoding

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The increasing demand for high-definition video services has led to a significant increase in data volume, necessitating improved video compression rates, with existing standards like HEVC showing limitations.

Innovation Solution

A method for partitioning video signals into tiles or slices, utilizing tile indices and difference information to enhance encoding and decoding efficiency, including techniques for parsing and encoding width and height information of slices.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of substance

If video compression rate is improved to handle high-definition video services, then data volume is reduced, but compression performance shows limitations with existing standards like HEVC

Engineering Contradiction:
Improvedata volumeVSAvoidcompression performance
Core Design Contradiction:
Loss of substanceVSReliability

Solution Approach 1:

The patent divides the picture into multiple tiles and slices, creating a hierarchical partitioning structure. This segmentation allows for more granular control of compression parameters across different regions, improving overall compression performance while handling high-definition video data volumes effectively

Inventive Principle:
Principle #1Segmentation

2Productivity

If picture is partitioned into multiple tiles and slices, then encoding/decoding efficiency is improved, but device complexity increases due to additional partitioning information processing

Engineering Contradiction:
Improveencoding/decoding efficiencyVSAvoidpartitioning information processing
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent performs picture partitioning into tiles and slices in advance during the encoding process, establishing a predefined structure that simplifies subsequent processing. The partitioning information is prepared and organized beforehand, reducing the computational burden during real-time encoding and decoding operations

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent introduces a two-dimensional partitioning approach by dividing the picture both horizontally (into slices) and vertically (into tiles), creating a grid-like structure. This multi-dimensional partitioning enables more efficient processing by organizing data in a structured manner that facilitates parallel processing and reduces complexity in subsequent operations

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

3Manufacturing precision

If rectangular slice partitioning is applied with explicit width and height information, then partitioning precision is improved, but data overhead increases due to additional signaling requirements

Engineering Contradiction:
Improvepartitioning precisionVSAvoiddata overhead
Core Design Contradiction:
Manufacturing precisionVSLoss of information

Solution Approach 1:

The patent applies different partitioning precision levels to different regions of the picture. Rectangular slices with explicit width and height information are used where high precision is needed, while other regions may use coarser partitioning. This localized approach maintains necessary precision for specific areas while reducing overall data overhead across the entire picture

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP4044593B1Video signal encoding/decoding method
Publication Date: 2026.03.11 APPLE INC
  • EP4044593B1 patent drawingFigure 1~2
  • EP4044593B1 patent drawingFigure 3~4
  • EP4044593B1 patent drawingFigure 5~6

AI summary

A video decoding method according to the present disclosure may include partitioning a current picture into a plurality of tiles, decoding partitioning information representing a slice type, wherein the partitioning information represents whether a rectangular slice is applied and decoding first width information and first height information of a slice when the partitioning information represents that the rectangular slice is applied.