Scalable Video Coding Weighted Prediction MCTF Update
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The scalable video coding standard, specifically the H.264 standard, does not provide clear guidance on handling weights in motion compensated temporal filtering (MCTF) update steps when weighted prediction is used in the prediction step, leading to inefficiencies and complexities in both encoding and decoding processes.
Innovation Solution
A scalable video encoder and decoder that perform motion compensated temporal filtering (MCTF) using a first weighting factor applied to a first reference picture, while utilizing a default update weighting factor applied to a second reference picture for the corresponding MCTF update step, where the default update weighting factor is derived from the first weighting factor used in the prediction step.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If weighted prediction is used in the MCTF prediction step, then coding efficiency is improved, but the complexity of handling weights in the MCTF update step increases
Solution Approach 1:
The patent changes the parameters used in the update step by applying inverse weighting factors (1/w0 and 1/w1) instead of the original weighting factors. This parameter transformation allows the update step to properly compensate for the weighted prediction applied in the analysis step, resolving the complexity issue while maintaining coding efficiency gains from weighted prediction
Solution Approach 2:
The patent implements a feedback mechanism where the weighting factors used in the prediction step are carried forward and inverted for use in the update step. This feedback loop ensures that the update operation correctly reverses the weighting effect, allowing the system to maintain both coding efficiency and manageable complexity through systematic parameter reuse
2Manufacturing precision
If separate weighting factors are used for prediction and update steps, then coding precision is improved, but device complexity increases
Solution Approach 1:
The patent makes the weighting factor calculation mechanism universal by using the same inverse weighting factors for both prediction and update steps. The apparatus computes weighting factors once and reuses them (inverted) in both phases, eliminating the need for separate complex weighting mechanisms while maintaining coding precision through appropriate parameter application in each step
3Device complexity
If default update weighting factors are used without derivation, then device complexity is reduced, but coding efficiency is lost
Solution Approach 1:
The patent performs preliminary action by calculating the weighting factors during the prediction step, and then reusing these pre-calculated factors (inverted) in the update step. This preliminary calculation avoids redundant computations and maintains coding efficiency while keeping the update step relatively simple through parameter reuse rather than recalculation
Data Source
AI summary
There are provided scalable video encoders, scalable video decoders, and corresponding methods for scalable video encoding and decoding. A scalable video encoder includes an encoder for encoding a block in a picture by performing a motion compensated temporal filtering (MCTF) prediction step for the block using a first weighting factor applied to a first reference picture, while using a different default update weighting factor applied to a second reference picture for a corresponding MCTF update step for the block.