Subpicture Signaling Scale Factor Adaptation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing method for signaling subpicture information in VVC is limited, as it only works effectively when the picture width and height are multiples of the subpicture unit, significantly reducing its applicability to various picture sizes and layouts.
Innovation Solution
A method is introduced that computes an initial width and height for subpictures by multiplying decoded scale factor values with the subpicture width and height values, adjusting these values if they exceed the picture dimensions, allowing for flexible application across different subpicture layouts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the existing method for signaling subpicture information is used, then the signaling is simple and direct, but it only works effectively when the picture width and height are multiples of the subpicture unit, significantly reducing its applicability
Solution Approach 1:
The patent introduces scale factor values as new parameters that allow the subpicture width and height to be scaled relative to the picture dimensions. This enables the signaling method to adapt to various picture sizes by changing the scale factor parameter, rather than requiring the picture dimensions to be multiples of subpicture units. The scale factor acts as a flexible parameter that bridges the gap between fixed subpicture units and variable picture sizes.
2Adaptability or versatility
If scale factor values are introduced to enable flexible subpicture layouts, then the applicability to non-multiple unit scenarios is enhanced, but the signaling method becomes more complex
Solution Approach 1:
The scale factor values are decoded and applied in advance during the initialization phase, before actual subpicture positioning and sizing operations. This preliminary calculation of scaled positions and dimensions simplifies subsequent processing steps, as the complex scaling logic is executed once upfront rather than repeatedly during decoding operations.
Data Source
AI summary
There is provided a method for decoding a position and a size for a subpicture, SP, in a picture from a bitstream. The method comprises decoding a coding tree unit, CTU, size from a first syntax element, S1, in the bitstream. The method comprises obtaining a scale factor value, F, wherein F is larger than (1). The method further comprises deriving a scaled position value for the subpicture SP, wherein deriving the scaled position value comprises: i) obtaining a position value based on information in the bitstream and ii) setting the scaled position value equal to the product of the position value and F. The method comprises deriving a size of the subpicture based on the scaled position value.