Hierarchical Coding Unit Pattern for Video Decoding Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video codecs face inefficiencies in encoding and decoding high-resolution video content due to limitations in encoding methods based on fixed macroblock sizes and coded block pattern information, which hinders the effective transmission and decoding of high-quality video.
Innovation Solution
A video decoding apparatus and method that utilize coding unit pattern information based on hierarchically structured coding units, allowing for encoding and decoding in units greater than macroblocks, and incorporating multiple hierarchical transformation units in a tree structure, enhancing encoding/decoding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If video is encoded according to a limited encoding method based on a macroblock having a predetermined size, then the encoding process is simple, but the encoding efficiency for high-resolution video content is insufficient
Solution Approach 1:
The video content is divided into coding units with hierarchical structures, where each coding unit can be further divided into transformation units. This segmentation allows flexible adaptation to different video content characteristics and resolutions, improving encoding efficiency while maintaining manageable complexity through systematic organization.
Solution Approach 2:
The coding unit size and transformation unit size are made dynamic rather than fixed. The encoder can select different coding unit sizes and transformation unit configurations based on the local video content characteristics, allowing the encoding method to adapt dynamically to varying content requirements for optimal efficiency.
2Manufacturing precision
If coded block pattern information is encoded in units of macroblocks, then the encoding structure is straightforward, but the encoding precision for high-quality video transmission is insufficient
Solution Approach 1:
The macroblock is segmented into smaller coding units and transformation units, allowing coded block pattern information to be encoded at a finer granularity. This segmentation enables more precise representation of video content characteristics, improving encoding precision for high-quality video transmission.
Solution Approach 2:
A hierarchical depth dimension is introduced to the coding unit structure, creating a multi-level organization system. This additional dimension allows for more detailed control and representation of video content at different levels of abstraction, enhancing encoding precision while providing a structured approach to manage complexity.
3Productivity
If coding unit pattern information is encoded in variously sized data units with hierarchical transformation units, then the encoding efficiency is improved, but the complexity of encoding and decoding processes increases
Solution Approach 1:
The coding units are organized in a nested hierarchical structure where larger coding units contain smaller transformation units, which themselves may contain further subdivided units. This nesting approach allows efficient representation of video content at multiple scales while providing a systematic framework that simplifies the management of complexity through recursive patterns.
Solution Approach 2:
The hierarchical coding unit structure serves multiple functions simultaneously: it enables flexible content adaptation, provides organized information representation, facilitates efficient transmission, and supports systematic decoding. This multi-functionality improves encoding efficiency while avoiding the need for separate specialized structures for each function.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
A method and apparatus for decoding video and a method and apparatus for encoding video are provided. The method for decoding video includes: receiving and parsing a bitstream of encoded video; and decoding encoded image data for maximum coding unit, based on information regarding the coded depth of the maximum coding unit, information regarding encoding mode, and coding unit pattern information.