Video Block Decoding Motion Vector Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional video decoding methods face memory bandwidth over-consumption and decreased picture accuracy when decoding bi-directionally predictive pictures, as they often rely on referencing either a previous or future video picture for motion compensation, which may not guarantee higher accuracy.
Innovation Solution
A method for decoding video blocks that selects N motion vectors from M motion vectors, utilizing intra pictures for higher accuracy and reducing memory bandwidth consumption by determining and using the most accurate motion vector, either directly from an intra picture or by calculating and comparing vectors from multiple reference areas.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If conventional schemes reserve only one motion vector corresponding to either previous or future video block, then memory bandwidth consumption is greatly alleviated, but decoding accuracy and final video quality are decreased
Solution Approach 1:
The patent applies partial action by selectively using only the necessary motion vectors based on picture type. For intra pictures, only one motion vector is used instead of two, and for bi-directional pictures, motion vectors are conditionally selected based on time interval comparisons. This partial usage reduces memory bandwidth consumption while maintaining sufficient decoding accuracy for the specific picture type.
Solution Approach 2:
The patent changes the parameter of motion vector selection by introducing time interval comparison as a decision criterion. Instead of always using both motion vectors or randomly selecting one, the system dynamically adjusts which motion vector to use based on the calculated time intervals between reference blocks and current block, optimizing both bandwidth usage and accuracy.
2Device complexity
If conventional schemes compare time intervals to select reference video block, then decoding process is simplified, but picture accuracy is not guaranteed because shorter time interval does not ensure higher accuracy
Solution Approach 1:
The patent introduces a feedback mechanism where the system calculates time intervals, compares them, and uses the comparison result to select the appropriate motion vector. This feedback loop ensures that the selection is based on objective criteria rather than simple assumptions, improving picture accuracy while maintaining a relatively simple decoding process through systematic decision-making.
Solution Approach 2:
The patent performs preliminary calculation of time intervals before selecting the motion vector. By pre-calculating these intervals and establishing comparison criteria in advance, the system prepares the necessary information beforehand, making the actual selection process efficient and accurate without adding significant complexity to the overall decoding flow.
Data Source
AI summary
A method for decoding a video block in a video picture includes: selecting N motion vectors from M motion vector, wherein the M motion vectors corresponding to at least a video block of at least a reference area; and decoding the video block according to the N motion vectors.