Video Inter Prediction Merge Offset for Motion Vector Refinement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing demand for high-definition video services has outpaced the compression capabilities of existing video coding standards like HEVC, leading to inefficiencies in data management and processing.
Innovation Solution
A method for deriving and refining motion information using a motion information table, including parsing flags for merge modes and applying merge offset encoding, and updating motion information based on block positions within a merge processing region to enhance inter-prediction efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of substance
If HEVC video compression standard is used, then compression performance is improved (about twice of H.264/AVC), but data amount still increases greatly with high-definition video services
Solution Approach 1:
The video signal is divided into multiple blocks, and each block is processed independently using motion compensation and prediction techniques. This segmentation allows for more efficient compression by handling smaller regions with specific motion characteristics, reducing the overall data amount while maintaining quality.
Solution Approach 2:
The patent employs advanced motion vector precision (1/8-pel or 1/16-pel) and multiple reference picture lists to change the parameters of motion estimation. This improves prediction accuracy, allowing for better compression ratios without significantly increasing data amount by more efficiently representing motion information.
2Productivity
If motion information table is updated with current block motion information, then inter-prediction efficiency is improved, but processing complexity increases
Solution Approach 1:
The motion information of the current block is pre-calculated and stored in the motion information table before it is needed for predicting other blocks. This preliminary action allows subsequent blocks to quickly access and use this motion information, improving inter-prediction efficiency without adding significant processing complexity during the main decoding process.
Solution Approach 2:
Instead of recalculating motion information for each block, the patent copies motion information from the motion information table (which contains motion vectors of previously decoded blocks) to predict motion of current blocks. This copying approach significantly reduces processing complexity while maintaining high inter-prediction efficiency.
3Device complexity
If merge mode is used for inter prediction, then encoding complexity is reduced, but prediction accuracy may be limited
Solution Approach 1:
The motion information table acts as an intermediary that stores motion information from multiple sources (spatial neighbors, temporal references). When merge mode is used, this intermediary provides additional motion candidates beyond simple spatial neighbors, enhancing prediction accuracy while keeping encoding complexity low through flag-based selection.
Solution Approach 2:
The merge mode is enhanced to serve multiple functions by incorporating motion information from both spatial and temporal references through the motion information table. This multi-functional approach allows a single encoding process to achieve both low complexity and high prediction accuracy by selecting from diverse motion candidates.
Data Source
AI summary
A method for decoding a video, according to the present invention, may comprise the steps of: parsing a first flag indicating whether inter prediction on the basis of a merge mode is applied to a current block; if the first flag is true, parsing a second flag indicating whether a regular merge mode or a merge offset encoding mode is applied to the current block; and if the second flag is true, parsing a third flag indicating whether the merge offset encoding mode is applied to the current block.