Transformed Reference Picture Insertion for Video Motion Compensation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image encoding/decoding methods face challenges in improving compression efficiency, particularly due to redundant information during prediction and the inability to effectively handle non-linear motions in video data.
Innovation Solution
The method involves creating transformed reference pictures by applying transformation information to reference pictures, determining their utilization, and incorporating them into the reference picture list for motion compensation, which reduces redundant operations and enhances prediction accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional motion compensation using original reference pictures is used, then the encoding/decoding process is simple, but compression efficiency is insufficient due to redundant information and inability to handle non-linear motions
Solution Approach 1:
The patent applies transformation information to reference pictures in advance to create transformed reference pictures before motion compensation. This preliminary transformation prepares the reference pictures to better match the current picture's non-linear motion characteristics, improving prediction accuracy and compression efficiency without adding complex real-time processing during decoding
Solution Approach 2:
The patent transforms reference pictures by applying transformation parameters (such as perspective transformation matrices) to modify their geometric properties. This parameter-based transformation allows the reference pictures to adapt to non-linear motions while maintaining a systematic and manageable encoding approach
2Measurement precision
If transformed reference pictures are created and inserted into reference picture lists, then prediction accuracy improves for non-linear motions, but the reference picture list management becomes more complex
Solution Approach 1:
The patent segments the reference picture list into multiple lists (e.g., List 0 and List 1), with transformed reference pictures inserted into specific lists based on their utility. This segmentation organizes the complex reference picture management into manageable groups, allowing the decoder to efficiently select appropriate reference pictures without overwhelming complexity
Solution Approach 2:
The patent introduces transformed reference pictures as intermediary elements between the original reference pictures and the current picture being encoded/decoded. These transformed pictures serve as a bridge that better represents non-linear motion relationships, improving prediction accuracy while the transformation information acts as a mediator that can be efficiently transmitted and applied
3Productivity
If transformation information is applied to all reference pictures, then all reference pictures can benefit from improved prediction, but redundant processing occurs for duplicate reference pictures
Solution Approach 1:
The patent merges duplicate reference pictures by identifying when multiple reference pictures are identical and applying transformation information only once. This consolidation eliminates redundant processing operations while ensuring that all references to duplicate pictures use the same transformed version, maintaining compression efficiency without wasted computational effort
Data Source
AI summary
The present invention relates to an image encoding/decoding method and apparatus, A method of decoding an image according to an exemplary embodiment of the present invention comprises: decoding transformation information; creating at least one transformed reference picture by applying the transformation information to at least one reference picture organized in a reference picture list; inserting the at least one transformed reference picture into the reference picture list; and performing motion compensation by using the reference picture list.