Video Encoder Intra-Refresh Block Prediction Restriction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video encoding schemes face challenges in reducing delay and efficiently encoding video data, particularly in real-time communication, where conventional methods may introduce delays due to picture rearrangement and require large buffers, and the intra-refresh scheme can lead to propagation of information from unrefreshed regions to refreshed regions.
Innovation Solution
A video encoder that employs an intra-refresh scheme with a restriction block identification unit, coding mode determination unit, prediction coding unit, and entropy coding unit to determine valid prediction modes and block sizes, ensuring that prediction blocks are generated using only information from refreshed regions, thereby preventing the use of unrefreshed region information and optimizing coding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If inter-picture prediction coding is used to achieve high compression ratio, then coding efficiency is improved, but delay increases and large buffers are required due to picture rearrangement
Solution Approach 1:
The picture is divided into multiple slices, and each slice is further divided into refresh blocks. This segmentation allows the encoder to process and transmit slices independently, reducing the need for large buffers and minimizing delay while maintaining compression efficiency.
Solution Approach 2:
The encoder performs intra-refresh by periodically refreshing specific blocks with intra-picture prediction before they are used as reference for inter-picture prediction. This preliminary action ensures that reference blocks are up-to-date, reducing delay and buffer requirements while maintaining high compression ratios.
2Loss of time
If intra-refresh scheme is used to reduce delay and buffer size, then delay and buffer requirements are reduced, but information from unrefreshed regions may propagate to refreshed regions
Solution Approach 1:
Different prediction modes are applied to different regions of the picture. Refresh blocks use intra-picture prediction to ensure they are self-contained and do not reference unrefreshed regions, while other blocks can use inter-picture prediction for better compression. This local differentiation prevents information propagation from unrefreshed to refreshed regions.
Solution Approach 2:
The encoder dynamically determines which blocks are refresh blocks based on slice structure and prediction mode selection. This dynamic approach allows flexible control over information propagation while optimizing compression efficiency in different regions of the picture.
3Productivity
If multiple prediction modes are considered for each block size to optimize coding efficiency, then coding efficiency is improved, but device complexity increases
Solution Approach 1:
Instead of evaluating all possible prediction modes for all block sizes, the encoder performs rate-distortion optimization only for relevant combinations. For refresh blocks, only intra-picture prediction modes are considered, while for other blocks, inter-picture prediction modes are evaluated. This partial evaluation reduces complexity while maintaining coding efficiency.
Data Source
AI summary
A video encoder includes: a restriction block identification unit that determines a valid prediction mode with a reference range not overlapping an unrefreshed region, for each size of a first sub-block, which is a unit for generating a prediction block, when a coding-target block in a refreshed region is to be encoded by intra prediction coding; and a coding mode determination unit that identifies a combination of the size of the first sub-block, the size of a second sub-block, which is a unit for application of a prediction mode, and a valid prediction mode with the smallest estimation value of the amount of code for encoding a third block, which is a unit for application of an intra prediction coding mode, as a combination to be used for intra prediction coding of the third sub-block.