Adaptive Weighted Prediction for Low-Latency Video Block Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies face challenges in efficiently compressing video data while maintaining decoder efficiency and accuracy, particularly in handling block sizes that require different levels of precision in weighted prediction.
Innovation Solution
Derive offset values and scaling factors for weighted prediction using partial neighboring samples based on block size thresholds, reducing decoder latency and signaling overhead, and implementing block-level adaptive weighted prediction (BAWP) to improve efficiency and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If all neighboring samples are used to derive offset values and scaling factors, then prediction accuracy is improved, but decoder latency increases
Solution Approach 1:
The patent applies partial action by using only a subset of neighboring samples (specifically, only those in the top and left regions) to derive offset values and scaling factors, rather than using all neighboring samples. This partial approach reduces the computational burden and latency while maintaining sufficient prediction accuracy for most cases, especially for smaller block sizes where the full set of samples would be required.
2Measurement precision
If scaling factor is signaled for all block sizes, then prediction accuracy is maintained, but signaling overhead increases
Solution Approach 1:
The patent applies local quality by differentiating the treatment of scaling factors based on block size. For smaller blocks (e.g., 4x4, 8x8), the scaling factor is derived implicitly without signaling, while for larger blocks (e.g., 16x16 and above), the scaling factor is explicitly signaled. This localized differentiation optimizes the balance between prediction accuracy and signaling overhead for different block sizes.
Solution Approach 2:
The patent changes the parameter of scaling factor handling based on block size thresholds. The system transitions from implicit derivation (for small blocks) to explicit signaling (for large blocks), optimizing the trade-off between accuracy and overhead dynamically based on the block dimensions being processed.
3Productivity
If block level adaptive weighted prediction is applied to all blocks, then coding efficiency is improved, but device complexity increases
Solution Approach 1:
The patent applies local quality by selectively applying block level adaptive weighted prediction (BAWP) only to specific blocks that meet certain criteria (e.g., larger block sizes or blocks with certain characteristics), rather than uniformly applying it to all blocks. This selective application maintains coding efficiency improvements where beneficial while reducing unnecessary complexity in cases where simple prediction suffices.
Data Source
AI summary
An example method of video coding includes receiving a video bitstream that includes a plurality of blocks. The method also includes obtaining a prediction sample for a current block of the plurality of blocks, and obtaining a scaling factor for the current block. The method further includes deriving an offset value for the current block. When a block size of the current block is less than a threshold, the offset value is derived based on a set of neighboring samples for the current block. When the block size of the current block is greater than the threshold, the offset value is derived based on only a subset of the set of neighboring samples. The method also includes adjusting the prediction sample using a linear equation with the scaling factor and the offset value and reconstructing the current block using the adjusted prediction sample.