Multi-Directional Motion Vector Prediction for Image Encoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional motion compensation techniques in image encoding/decoding are limited by the use of only spatial, temporal, and zero motion vector candidates, and uni-directional or bi-directional predictions, which restricts encoding/decoding efficiency for high-resolution and high-quality images.
Innovation Solution
The method generates multiple motion vector candidate lists for inter-prediction, including spatial, temporal, and combined motion vectors, and uses uni-directional, bi-directional, tri-directional, and quad-directional predictions to derive multiple motion vectors and prediction blocks, with weighted summation to determine the final prediction block.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If only spatial, temporal, and zero motion vector candidates are used in conventional motion compensation, then the device complexity is low, but the encoding/decoding efficiency is insufficient for high-resolution images
Solution Approach 1:
The motion vector candidate list is segmented into multiple directional categories (spatial, temporal, combined, and zero motion vector candidates). Each segment serves a specific prediction purpose, allowing the system to selectively use appropriate candidates for different block types and motion patterns, thereby improving efficiency without overwhelming complexity
Solution Approach 2:
The patent extends the traditional motion compensation approach by adding combined motion vector candidates that integrate both spatial and temporal information. This creates an additional dimension in the candidate selection space, enabling more accurate predictions for complex motion patterns in high-resolution images while maintaining a structured candidate list framework
2Measurement precision
If only uni-directional and bi-directional predictions are used, then the processing complexity is manageable, but the prediction accuracy is limited for complex motion patterns
Solution Approach 1:
The patent implements dynamic prediction by allowing the selection of different prediction directions (uni-directional, bi-directional, tri-directional, quad-directional) based on the specific characteristics of each video block. The system adapts the prediction complexity to match the actual motion complexity, using more directional candidates when needed while maintaining simplicity for straightforward cases
Solution Approach 2:
The patent creates a composite prediction approach by combining multiple motion vector candidate types (spatial, temporal, combined, zero) across multiple directions. This composite structure allows the system to leverage the strengths of different candidate types and prediction directions, achieving superior prediction accuracy for complex motion patterns while maintaining a systematic framework
Data Source
AI summary
The present invention relates to a method of performing motion compensation by using motion vector prediction. To this end, a method of decoding an image may include: generating multiple motion vector candidate lists according to an inter-prediction direction of a current block; deriving multiple motion vectors for the current block by using the multiple motion vector candidate lists; determining multiple prediction blocks for the current block by using the multiple motion vectors; and obtaining a final prediction block for the current block based on the multiple prediction blocks.