Video Coding Constraint Flag for Prediction Unit Complexity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increased degree of freedom in splitting Coding Tree Units (CTUs) using BT split in video coding schemes leads to reduced PU size and increased coding amounts, necessitating a method to maintain large block prediction while reducing coding complexity.
Innovation Solution
An image decoding and coding apparatus that employs a constraint flag to control the decoding and coding of prediction units, constraining decoding/encoding in certain nodes to manage complexity and reduce coding amounts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If BT split is introduced in addition to QT split to improve the degree of freedom of splitting CTU, then the degree of freedom of splitting into PU and TU is improved, but the PU size is reduced and the coding amount of PU increases
Solution Approach 1:
The patent applies partial action by selectively constraining PU decoding only in specific coding nodes where it is most beneficial, rather than uniformly across all nodes. The constraint flag mechanism allows the system to apply PU constraint selectively in certain coding trees while maintaining normal PU decoding in others, thus achieving partial optimization that reduces overall coding amount without completely sacrificing splitting flexibility.
2Adaptability or versatility
If BT split is introduced in addition to QT split to improve the degree of freedom of splitting CTU, then the degree of freedom of splitting into PU and TU is improved, but the complexity of coding/decoding increases
Solution Approach 1:
The patent implements preliminary action by pre-defining constraint flags in the coding tree structure that indicate whether PU constraint should be applied in specific nodes. This preliminary configuration allows the decoder to anticipate and prepare for constrained decoding regions, avoiding complex real-time decisions during decoding and thereby reducing overall coding/decoding complexity while maintaining splitting flexibility.
3Quantity of substance
If PU constraint is applied to reduce coding amount, then the coding amount of PU is reduced, but the flexibility in decoding prediction units is reduced
Solution Approach 1:
The patent applies dynamics by making the PU constraint flexible and adaptive through the constraint flag mechanism. The constraint is not fixed or static but can be dynamically enabled or disabled in different coding nodes based on the flag values. This allows the system to adaptively apply constraints where beneficial while maintaining flexibility in other regions, thus reducing coding amount without permanently sacrificing decoding flexibility.
Data Source
AI summary
An image decoding apparatus (31) includes a CT information decoding unit (10) configured, in each coding tree, to decode a constraint flag indicating whether or not to constrain decoding a prediction unit, and a CU decoding unit (20) configured, in the coding tree, to decode a prediction unit in a coding node to be decoded first and to not decode a prediction unit in another coding node, in a case that the constraint flag indicates constraint of decoding the prediction unit. According to the present invention, it is possible to reduce a coding amount of PU and complexity of coding/decoding an image while maintaining a split with a high degree of freedom by CT split (QT split and BT split).