Inter Prediction Motion Adjustment for Lower-Bit Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video compression technologies face inefficiencies in block division structures and increased bit usage due to various inter prediction modes and memory bandwidth consumption, particularly in high-resolution and high-definition video encoding and decoding processes.
Innovation Solution
The method and apparatus introduce adaptive motion information adjustment for bidirectional prediction based on predefined inter modes, utilizing merge and affine modes to derive motion information from candidate lists and affine candidates, and hierarchical picture division to optimize memory usage and prediction efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If various block division structures (QTBT, MTT) and multiple inter prediction modes are used to improve video coding efficiency, then prediction accuracy is improved, but the number of bits expressing coding information increases significantly
Solution Approach 1:
The patent applies parameter changes by dynamically adjusting the block division structure parameters (switching between QTBT and MTT) and inter prediction mode parameters based on picture characteristics. This allows the encoder to select the most efficient combination of block sizes and prediction modes for different regions, improving prediction accuracy while avoiding the use of all possible modes and sizes, thereby reducing the bits required to express coding information.
2Measurement precision
If multiple motion vectors are used for bidirectional prediction to improve prediction quality, then inter prediction efficiency is improved, but memory bandwidth consumption increases
Solution Approach 1:
The patent implements partial action by using multiple motion vectors for bidirectional prediction only when necessary, rather than applying it universally to all blocks. The encoder evaluates picture characteristics and selectively applies bidirectional prediction with multiple motion vectors to regions that benefit most from it, while using simpler unidirectional prediction for other regions. This reduces memory bandwidth consumption while maintaining prediction quality where it matters most.
3Productivity
If one picture is divided into multiple blocks with various coding modes to improve compression efficiency, then video coding efficiency is improved, but device complexity increases
Solution Approach 1:
The patent applies dynamics by making the block division structure adaptive rather than static. The encoder dynamically selects between QTBT and MTT structures, and adjusts block sizes and prediction modes based on local picture characteristics. This dynamic adaptation allows the system to achieve high coding efficiency by matching the block structure to the content, while avoiding the complexity of implementing all possible block division types simultaneously in all regions.
Data Source
AI summary
In a method and a device for encoding or decoding an image according to the present invention, motion information for bidirectional prediction of a current block may be derived on the basis of an inter mode previously defined in the device for encoding or decoding an image, and inter prediction may be performed on the current block on the basis of the motion information, wherein the motion information for bidirectional prediction is adjusted to be motion information for unidirectional prediction according to the predefined inter mode.