Image Encoding Sub-Block Transform and Merge Candidate Deduplication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video compression technologies face challenges in efficiently encoding and decoding high-resolution and ultra-high resolution stereoscopic image contents, particularly in improving prediction methods, transform techniques, and motion compensation processes.
Innovation Solution
The proposed method and device for image encoding/decoding generate prediction blocks based on predefined prediction modes, residual blocks based on predetermined transform types, and reconstruct current blocks using these blocks. This includes intra prediction using intra reference samples, merge mode prediction with motion compensation, and multi-type based transforms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional video compression technology is used, then encoding/decoding can be performed, but encoding efficiency is insufficient for high-resolution and ultra-high resolution stereoscopic image contents
Solution Approach 1:
The current block is divided into multiple sub-blocks for independent transform processing. Different transform types can be applied to different sub-blocks based on their local characteristics, improving compression efficiency for high-resolution content without requiring a complete redesign of the encoding framework
Solution Approach 2:
The transform type is dynamically selected based on the characteristics of each sub-block rather than applying a uniform transform to the entire block. This adaptive approach allows the encoding system to optimize for local variations in high-resolution stereoscopic content, improving overall encoding efficiency
2Productivity
If a single transform type is used for the entire current block, then encoding is simple, but encoding efficiency is limited
Solution Approach 1:
The current block is segmented into multiple sub-blocks, each of which can undergo different transform types. This segmentation allows the system to achieve improved encoding efficiency through selective transform application while keeping the complexity manageable by processing smaller sub-blocks independently
Solution Approach 2:
Different transform types are applied to different sub-blocks based on their local characteristics. This local optimization approach improves overall encoding efficiency by adapting to local variations in the image content without requiring complex global optimization
3Adaptability or versatility
If merge candidates from motion information list are overlapped with spatial merge candidates, then more candidates are available, but redundancy increases
Solution Approach 1:
The patent extracts and removes duplicate merge candidates from the motion information list that overlap with spatial merge candidates. This extraction of redundant information maintains the versatility of having multiple candidate types while eliminating the waste of storing duplicate motion information
Solution Approach 2:
Redundant merge candidates are discarded from the motion information list to avoid duplication with spatial merge candidates. The essential motion information is preserved through the spatial candidates, while the discarded temporal candidates prevent unnecessary bitstream overhead
Data Source
AI summary
An image encoding/decoding method and apparatus according to the present invention may generate a prediction block of a current block on the basis of a pre-defined prediction mode, generate a residual block of the current block on the basis of a predetermined transformation type, and reconstruct the current block on the basis of the prediction block and the residual block.