Decoder-Side Motion Compensation Using Sub-Block SAD Refinement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies face challenges in efficiently compressing high-resolution images and 3D video content, particularly in adapting motion information for improved encoding and decoding efficiency.
Innovation Solution
The method involves performing motion modification in units of sub-blocks, determining a search region, calculating SAD candidates, and deriving delta motion information to adaptively modify motion information based on pre-generated information and threshold values, considering block size, distance, prediction mode, and resolution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If motion modification is performed using traditional methods, then decoding complexity is reduced, but compression efficiency deteriorates
Solution Approach 1:
The decoder performs motion modification using SAD candidates derived from the bitstream and pre-generated motion information, making the decoder self-sufficient in generating accurate motion vectors without requiring additional encoder computations or complex inter-encoder/decoder communication
Solution Approach 2:
The patent changes the parameter representation by using SAD (Sum of Absolute Differences) candidates as a basis for deriving motion vector differences, transforming the traditional motion compensation approach into a more efficient parameter-driven method that improves compression while maintaining manageable decoding complexity
2Productivity
If high-resolution video is compressed using conventional standards, then video quality is maintained, but compression ratio deteriorates
Solution Approach 1:
The patent segments the motion compensation process into distinct components: SAD candidate generation, motion vector difference derivation, and motion vector refinement. This segmentation allows for more precise control over compression at different stages while maintaining overall video quality through systematic optimization
Solution Approach 2:
By introducing SAD-based motion vector derivation as an intermediate parameter, the patent enables more efficient representation of motion information in high-resolution video, achieving better compression ratios while preserving visual quality through improved motion estimation accuracy
3Measurement precision
If motion information is modified with high precision, then prediction accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent performs preliminary action by pre-generating motion information at the encoder side and using it to guide the SAD candidate selection process at the decoder. This preliminary preparation enables high-precision motion vector derivation without requiring the decoder to perform exhaustive searches, thus balancing accuracy with computational efficiency
Solution Approach 2:
The patent applies local quality by focusing computational resources on generating SAD candidates for specific regions of interest in the video frame. Rather than uniformly processing all blocks, the method adapts the level of motion modification based on local characteristics, achieving high prediction accuracy where needed while reducing overall computational complexity
Data Source
AI summary
A video coding method and device, according to the present invention, determine whether motion compensation is performed by sub-block unit, determine a search area for motion compensation of a current block, calculate a plurality of SAD candidates with respect to the search area, derive delta motion information of the current block on the basis of the plurality of SAD candidates, and can compensate for motion information of the current block by using pre-generated motion information and the delta motion information of the current block.