Multi-Hypothesis Video Prediction for Complex Motion Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding systems, such as VVC, face challenges in efficiently handling complex video content with diverse motion patterns, leading to increased computational complexity and reduced coding efficiency.
Innovation Solution
The implementation of multiple hypothesis prediction methods, including extended merge prediction, affine motion compensation, and adaptive motion vector resolution, to enhance coding efficiency by refining motion parameter signaling and prediction processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple hypothesis prediction methods are implemented to handle complex motion patterns, then coding efficiency is improved, but computational complexity increases
Solution Approach 1:
The prediction process is divided into multiple independent hypotheses (first hypothesis using conventional bi-prediction, second hypothesis using alternative motion parameters). Each hypothesis is evaluated separately and combined through weighted averaging, allowing parallel computation without sequential dependencies. This segmentation enables efficient handling of complex motion patterns while maintaining manageable computational load at each stage.
Solution Approach 2:
The patent implements a limited number of prediction hypotheses (typically 2-3) rather than exhaustively searching all possible motion parameters. By selecting only the most promising hypotheses based on preliminary criteria and combining them with weighted averaging, the system achieves improved coding efficiency without the exponential computational cost of complete motion parameter search.
2Device complexity
If conventional single hypothesis prediction is used, then computational complexity is reduced, but coding efficiency deteriorates for complex video content
Solution Approach 1:
The patent merges multiple prediction hypotheses (conventional bi-prediction from reference pictures and alternative prediction from current picture or other reference pictures) through weighted averaging. This combination allows the system to leverage the simplicity of conventional methods while incorporating additional prediction information, achieving improved coding efficiency for complex video content without requiring a complete overhaul of the prediction architecture.
Solution Approach 2:
The multi-hypothesis prediction framework is designed to work with various video content types and motion patterns. By maintaining multiple prediction paths (conventional and alternative) that can be adaptively combined, the system achieves universal applicability across different video sequences, handling both simple and complex motion patterns effectively without requiring content-specific customization.
Data Source
AI summary
A method for predictive coding operates by receiving input data associated with a current block, wherein the input data comprise pixel data for the current block to be encoded at an encoder side or encoded data associated with the current block to be decoded at a decoder side. The method then determines prediction members associated with an existing prediction, wherein each prediction member corresponds to one weighted sum of the existing prediction and at least one target prediction candidate from a group of prediction candidates, and wherein a weight and the at least one target prediction candidate are jointly decided by a joint index. Encoding or decoding of the current block is performed using a final prediction decided based on the joint index.