Motion Vector Predictor Redundancy Removal in Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In video coding systems, particularly in High-Efficiency Video Coding (HEVC), there is a challenge in reducing redundancy in motion vector predictors to enhance coding efficiency and prevent parsing errors caused by temporal motion prediction, which can lead to severe error propagation due to mismatched candidate sets between the encoder and decoder.
Innovation Solution
A method is introduced to identify and remove redundant motion vector predictor (MVP) candidates based on non-MV-value criteria, such as same Processing Units (PU), Motion Data Storage Reduction (MDSR) units, block merging, and motion IDs, without comparing MV values, to reduce complexity and improve performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If temporal motion prediction is used to reduce bitrate, then coding efficiency is improved, but parsing errors and error propagation occur due to mismatched candidate sets between encoder and decoder
Solution Approach 1:
The patent applies preliminary action by identifying and removing redundant MVP candidates from the candidate set before the actual motion vector prediction process. This preprocessing step ensures that both encoder and decoder operate with identical candidate sets, preventing mismatch errors during decoding while maintaining the benefits of temporal prediction for coding efficiency.
2Measurement precision
If MVP candidate sets are expanded to improve prediction accuracy, then coding performance is enhanced, but complexity of candidate set management and redundancy increases
Solution Approach 1:
The patent extracts and removes redundant MVP candidates from the candidate set based on specific redundancy criteria. This extraction process reduces the size and complexity of the candidate set while retaining the most predictive candidates, thereby simplifying candidate set management without significantly compromising prediction accuracy.
Solution Approach 2:
The patent discards redundant MVP candidates that provide duplicate or suboptimal prediction information. By removing these redundant candidates, the system reduces computational complexity and memory requirements while the essential prediction candidates are preserved for accurate motion vector prediction.
3Loss of substance
If motion vector coding is performed in low-bitrate applications, then bandwidth consumption is reduced, but the portion of compressed video data required for motion vectors becomes more significant
Solution Approach 1:
The patent uses spatial and temporal copying of motion vector information from neighboring blocks and reference pictures to generate MVP candidates. This copying approach allows the encoder to predict motion vectors without transmitting complete motion vector data, significantly reducing the proportion of compressed video data required for motion vectors in low-bitrate applications.
Data Source
AI summary
A method implemented in an apparatus for video coding a current block coded in an Inter, Merge, or Skip mode determines neighboring blocks of the current block, wherein a motion vector predictor (MVP) candidate set is derived from MVP candidates associated with the neighboring blocks. The method determines at least one redundant MVP candidate, if said MVP candidate is within a same PU (Prediction Unit) as another MVP candidate in the MVP candidate set. The method removes said at least one redundant MVP candidate from the MVP candidate set, and provides a modified MVP candidate set for determining a final MVP, wherein the modified MVP candidate set corresponds to the MVP candidate set with said at least one redundant MVP candidate removed. Finally, the method encodes or decodes the current block according to the final MVP.