Inter Prediction Motion Vector Coding to Reduce Video Redundancy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies face challenges in reducing redundancy and improving coding efficiency, particularly in inter prediction methods, which affect the performance of video compression.
Innovation Solution
The proposed inter prediction method involves obtaining a motion vector predictor and determining motion vector differences based on preset candidate length and direction information, using a set of candidate length or direction information to refine the motion vector target value for improved prediction accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a full set of candidate motion vector differences is used for prediction, then prediction accuracy is improved, but coding complexity and redundancy increase
Solution Approach 1:
The patent segments the motion vector difference information into two separate components: length information and direction information. This segmentation allows the encoder to process and encode these components independently, reducing the overall complexity while maintaining prediction accuracy. The motion vector difference is divided such that length candidates are selected from a first set and direction candidates from a second set, enabling separate candidate list construction and processing.
Solution Approach 2:
The patent transforms the traditional single-dimension motion vector difference representation into a two-dimensional structure by separating length and direction components. This dimensional change allows for more efficient candidate management where length and direction can be independently optimized and encoded, reducing redundancy in the candidate lists while preserving the complete motion information needed for accurate prediction.
2Productivity
If motion vector difference information is compressed more aggressively, then coding efficiency is improved, but prediction accuracy deteriorates
Solution Approach 1:
The patent changes the parameter representation by separating motion vector differences into length and direction parameters. This allows for selective precision allocation where length can be represented with coarser granularity while direction uses discrete angular values, optimizing the balance between compression efficiency and prediction accuracy. The candidate lists are constructed with specific numbers of length candidates and direction candidates, enabling controlled information retention.
Solution Approach 2:
The patent applies different quality levels to different components of the motion vector difference. The length component and direction component can have different candidate list sizes and precision levels tailored to their respective importance and statistical characteristics. This local quality approach ensures that critical information is preserved while less critical aspects are compressed more aggressively, maintaining overall prediction accuracy while improving coding efficiency.
Data Source
AI summary
An inter prediction method and apparatus, and a corresponding encoder and decoder are disclosed. The method includes: obtaining an index value of a length of a motion vector difference of the current picture block; determining target length information from a set of candidate length information based on the index value of the length; obtaining the motion vector difference of the current picture block based on the target length information; determining the motion vector target value of the current picture block based on the motion vector difference of the current picture block and the motion vector predictor of the current picture block; and obtaining a prediction block of the current picture block based on the motion vector target value of the current picture block.