IBC Virtual Buffer Segmentation for Video Coding Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding standards face challenges in efficiently managing virtual buffers for intra block copy (IBC) operations, particularly in handling large CTU sizes and ensuring proper reference sample availability across VPDU rows and columns.
Innovation Solution
The proposed solution involves implementing general virtual buffers for IBC, where the size and allocation of these buffers are dynamically determined based on VPDU, CTB, and CTU sizes. This includes constraints on block vectors and reference sample marking to ensure valid predictions and efficient memory usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If larger IBC virtual buffer sizes are used to improve prediction accuracy, then video coding efficiency is improved, but memory usage and device complexity increase
Solution Approach 1:
The IBC virtual buffer is segmented into multiple line buffers corresponding to different VPDU rows. Each line buffer stores reference samples for a specific row of VPDUs, allowing the system to manage large buffer sizes through structured organization. This segmentation enables efficient memory access patterns while maintaining the ability to handle large IBC reference regions without proportionally increasing complexity.
Solution Approach 2:
The patent introduces a new dimensional organization for the IBC virtual buffer by mapping 2D buffer access patterns to 1D line buffer structures. Instead of managing a single large 2D buffer, the system uses multiple 1D line buffers indexed by VPDU row position, transforming the memory management problem into a more tractable form that reduces computational overhead while preserving large buffer capacity.
2Manufacturing precision
If reference samples are marked as unavailable based on chroma block position to ensure prediction accuracy, then manufacturing precision is improved, but device complexity increases
Solution Approach 1:
The patent applies local quality by marking reference samples as unavailable based on their specific spatial relationship with chroma blocks. Instead of uniformly marking all reference samples, the system selectively marks only those samples that would cause prediction errors due to chroma block positioning. This localized approach maintains high prediction accuracy while minimizing the number of marked samples and associated complexity.
3Reliability
If block vectors are constrained to prevent crossing CTU rows to ensure proper reference sample availability, then reliability is improved, but productivity decreases
Solution Approach 1:
The patent implements dynamic block vector constraints that adapt to the current coding context. Rather than applying fixed constraints that always prevent CTU row crossing, the system dynamically determines whether crossing is permissible based on the availability of reference samples in the IBC virtual buffer. This dynamic approach maintains reliability by ensuring reference sample availability when possible while maximizing productivity by allowing crossings when safe.
Data Source
AI summary
A method of video processing is described. The method includes determining, for a conversion between a current video block of a video picture of a video and a coded representation of the video, whether or how to reset reference sample values of a reference region of the video picture used for predicting the current video block according to a rule; and performing the conversion based on the determining.