Merge Candidate Reordering for Higher-Efficiency Image Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
High-resolution and high-quality image data requires high-efficiency encoding/decoding techniques to reduce the increased costs of transmission and storage associated with conventional methods.
Innovation Solution
An image encoding/decoding method that utilizes a reconfigured merge candidate list, calculated using distortion values such as SAD, SATD, and MR-SAD, to improve compression efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional image encoding methods are used, then transmission and storage costs increase with higher resolution and quality, but compression efficiency remains insufficient
Solution Approach 1:
The patent dynamically reconfigures the merge candidate list by calculating distortion values (SAD, SATD, MR-SAD) for different merge candidates and rearranging them based on actual image content characteristics. This dynamic adaptation allows the encoder to optimize compression efficiency for each specific image block, resolving the contradiction between achieving high compression efficiency and managing transmission/storage costs.
Solution Approach 2:
The patent changes the parameter ordering in the merge candidate list based on calculated distortion values. By computing multiple distortion metrics (SAD for sum of absolute differences, SATD for sum of absolute transformed differences, MR-SAD for mean removed sum of absolute differences) and using these to reconfigure the candidate list, the system optimizes the balance between compression efficiency and resource consumption.
2Productivity
If inter-prediction is performed with a fixed merge candidate list, then the encoding process is simple, but compression efficiency is suboptimal
Solution Approach 1:
The patent performs preliminary calculations of distortion values (SAD, SATD, MR-SAD) for all merge candidates before finalizing the merge candidate list. This preliminary action allows the system to pre-determine the optimal candidate ordering based on image content, improving compression efficiency without significantly increasing the overall encoding complexity during the actual prediction process.
Solution Approach 2:
The patent implements feedback by calculating distortion values based on actual image blocks and using these feedback results to reconfigure the merge candidate list. The distortion calculations provide feedback about which merge candidates are most effective for each specific block, allowing the system to adapt the candidate list to maximize compression efficiency.
3Productivity
If distortion calculation is performed for all merge candidates, then compression efficiency improves, but processing time increases
Solution Approach 1:
The patent applies partial action by calculating distortion values for merge candidates in a structured manner, focusing computational resources on the most relevant candidates. By using multiple distortion metrics (SAD, SATD, MR-SAD) selectively and arranging candidates in order of relevance, the system achieves improved compression efficiency while managing processing time through targeted computation rather than exhaustive uniform processing.
Data Source
AI summary
An image encoding/decoding method may comprise deriving at least one merge candidate of a current block, generating an initial merge candidate list of the current block by using the derived at least one merge candidate and generating a reconfigured merge candidate list by using the initial merge candidate list.