Inter-Layer Prediction Rules for Mixed Coding Block Sizes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing Versatile Video Coding (VVC) standard does not support an improved 360° video coding scheme that combines inter-layer prediction (ILP) with subpictures and spatial scalability, leading to inefficiencies and unnecessary coding constraints.
Innovation Solution
Introduce multiple SPS flags to control reference picture resampling (RPR) and allow flexible use of inter-layer prediction, relax constraints on subpicture alignment, and enable negative scaling window offsets to support efficient 360° video coding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the existing VVC standard is used with strict constraints on reference picture resolution and subpicture alignment, then decoding complexity is reduced, but coding efficiency deteriorates due to inability to support improved 360° video coding schemes
Solution Approach 1:
The patent introduces dynamic control mechanisms through multiple syntax flags (res_change_in_clvs_allowed_flag, subpic_treated_as_pic_flag, collocated_view_pic_constraint_flag) that allow the coding system to adaptively switch between different operational modes. This enables the system to be strict when needed and flexible when beneficial, resolving the contradiction between coding efficiency and decoding complexity.
Solution Approach 2:
The patent changes key parameters by allowing reference picture resolution changes within coded layer video sequences, enabling subpictures to be treated as independent pictures for prediction purposes, and modifying constraints on collocated picture selection. These parameter changes enable improved 360° video coding schemes while maintaining backward compatibility through flag-controlled activation.
2Adaptability or versatility
If multiple syntax flags and relaxed constraints are introduced to support flexible inter-layer prediction and reference picture resampling, then adaptability improves, but standard complexity increases
Solution Approach 1:
The patent segments the control mechanism into multiple independent syntax flags, each governing a specific aspect of flexibility (resolution changes, subpicture treatment, collocated picture constraints). This segmentation allows the system to enable only the necessary flexibility for each coding scenario, avoiding the need to implement all possible flexibilities simultaneously, thus managing standard complexity while maintaining adaptability.
Solution Approach 2:
The introduced syntax flags and relaxed constraints create a universal framework that can support multiple coding schemes including improved 360° video coding, spatial scalability, and inter-layer prediction. This multi-functional design allows a single standardized mechanism to serve multiple purposes, improving adaptability without proportionally increasing complexity.
3Adaptability or versatility
If constraints on cross-layer alignment and subpicture combination are relaxed, then support for spatial scalability improves, but processing complexity increases
Solution Approach 1:
The patent applies dynamic control through the subpic_treated_as_pic_flag and collocated_view_pic_constraint_flag to manage cross-layer alignment and subpicture combination constraints. When spatial scalability is needed, these flags enable relaxed constraints; otherwise, standard constraints apply. This dynamic approach supports spatial scalability while managing processing complexity through conditional activation.
Data Source
AI summary
Several techniques for video encoding and video decoding are described. One example method includes performing a conversion between a video unit of a current picture of a video and a bitstream of the video according to a rule. The rule specifies that, for a given picture to be available as a reference picture whose motion information is used for the conversion, the given picture has a same coding tree unit size or a same coding tree block size as the current picture.