BIO Coding Unit Partitioning for Smaller Inter Prediction Buffers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In the VVC standard, partitioning PUs larger than VPDUs for inter prediction with BIO is challenging due to the need for large buffer resources and high computational costs, making it difficult to virtually partition PUs into vPUs for efficient BIO processing.
Innovation Solution
The processing unit is partitioned into smaller units, such as vPUs, equal to or smaller than the VPDU size, to determine the applicability of BIO, reducing the required buffer size and computational costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If BIO processing is applied to PUs larger than VPDUs, then encoding efficiency is improved, but buffer size and computational costs increase significantly
Solution Approach 1:
The patent divides a large PU into multiple smaller vPUs (virtual PUs), each of which can be processed independently with smaller buffer resources. The BIO processing is applied selectively to individual vPUs based on local SAD calculations, rather than requiring buffers for the entire large PU. This segmentation allows BIO to be applied efficiently to regions where it provides benefit while avoiding excessive buffer requirements.
2Productivity
If BIO processing is applied to PUs larger than VPDUs, then encoding efficiency is improved, but computational costs increase significantly
Solution Approach 1:
The patent applies BIO processing selectively to specific vPUs based on local SAD (Sum of Absolute Differences) calculations. Instead of applying BIO uniformly across the entire large PU, the method evaluates each vPU individually and applies BIO only where it provides significant encoding benefit. This local quality approach reduces overall computational costs while maintaining encoding efficiency in regions where BIO is most effective.
Solution Approach 2:
The patent implements partial BIO processing by applying the computationally intensive BIO algorithm only to selected vPUs rather than the entire PU. By using SAD thresholds to identify regions where BIO is beneficial, the method performs partial action on subsets of the data, reducing total computational costs while still achieving encoding efficiency improvements in the most suitable regions.
3Measurement precision
If SAD calculation is performed on entire PU to determine BIO applicability, then determination accuracy is improved, but processing complexity increases and virtual partitioning becomes difficult
Solution Approach 1:
The patent segments the SAD calculation process by computing SAD values for individual vPUs rather than calculating SAD for the entire large PU. This segmentation reduces processing complexity at each stage while maintaining sufficient determination accuracy for BIO applicability. The divided approach allows parallel processing and integrates smoothly with the virtual partitioning structure.
Data Source
AI summary
The present technology relates to image processing device and method that achieve a reduction in a buffer size.The image processing device partitions a unit of processing into partitioned processing units each of which corresponds to a VPDU size or is equal to or smaller than the VPDU size, the unit of processing being used for calculation of a cost that is used for determining whether or not to perform bidirectional prediction. The image processing device makes the determination by using the cost calculated based on the partitioned processing units. The present technology is applicable to encoding devices or decoding devices.