Image Decoding Merge Candidate List Reordering for Coding Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image coding technologies face inefficiencies in compressing and transmitting high-resolution, high-quality images, leading to increased transmission and storage costs due to the high amount of information required.
Innovation Solution
An image decoding method and device that construct a merge candidate list based on neighboring blocks, derive costs for merge candidates, and perform prediction using a modified merge candidate list to optimize coding efficiency by reordering and refining merge candidates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional image coding techniques are used for high-resolution images, then image quality is maintained, but transmission and storage costs increase due to high bit amounts
Solution Approach 1:
The current block is divided into multiple sub-blocks for separate motion prediction. Each sub-block can have its own motion vector derived from different merge candidates, allowing more precise local motion representation without uniformly increasing bits for the entire block. This segmentation enables differential coding where only motion vector differences are transmitted for each sub-block.
Solution Approach 2:
The patent dynamically constructs merge candidate lists based on spatial and temporal neighboring blocks, adapting the prediction approach to local image characteristics. The system selectively applies different prediction modes (merge, skip, motion compensation) based on motion complexity, allowing efficient bit allocation that matches actual image content requirements rather than using a fixed high-bitrate approach.
2Measurement precision
If more merge candidates are included in the merge candidate list, then prediction accuracy improves, but coding complexity increases
Solution Approach 1:
The patent applies different levels of merge candidate refinement to different regions based on their motion characteristics. For blocks with simple motion, a smaller set of merge candidates is sufficient. For blocks with complex motion patterns, the system dynamically adds more candidates from spatial and temporal neighboring blocks. This local adaptation optimizes prediction accuracy where needed while avoiding unnecessary complexity in simpler regions.
Solution Approach 2:
The system performs preliminary construction of merge candidate lists using available spatial and temporal neighboring blocks before final selection. By pre-organizing potential candidates and their associated costs, the system reduces the computational burden during actual encoding/decoding. The preliminary list construction allows rapid evaluation and selection of optimal candidates without exhaustive search.
3Productivity
If merge candidates are reordered based on cost, then coding efficiency improves, but processing time increases
Solution Approach 1:
The patent changes the ordering parameter of merge candidates from arbitrary or spatial order to cost-based ordering. By calculating prediction costs (such as SAD, SSD, or PSNR metrics) for each candidate and sorting accordingly, the system ensures that the most promising candidates are evaluated first. This parameter change enables early termination strategies where encoding can stop after evaluating a sufficient number of top candidates, reducing overall processing time while maintaining high coding efficiency.
Data Source
AI summary
An image decoding method that is performed by a decoding apparatus according to the present disclosure comprises the steps of: forming a merge candidate list based on neighbouring blocks of a current block; deriving costs with respect to merge candidates included in the merge candidate list; deriving a revised merge candidate list based on the costs with respect to the merge candidates; deriving movement information of the current block based on the revised merge candidate list; and performing prediction on the current block based on the movement information, wherein the neighbouring blocks include spatial neighbouring blocks and temporal neighbouring blocks.