Inter-Frame Prediction Modes for Block-Level Texture Preservation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video encoding and decoding technologies use a uniform predicted value derivation mode for all blocks in a picture frame, which fails to account for varying motion information and textures, leading to inefficiencies and reduced image quality, particularly in regions with sharp textures.
Innovation Solution
An inter prediction method that determines individual predicted value derivation modes for each block based on its motion information and position, allowing for flexible and accurate prediction by considering adjacent blocks and their motion information, and adjusting prediction modes to avoid texture blurring.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a uniform predicted value derivation mode is used for all blocks in a picture frame, then the device complexity is reduced and processing is simplified, but the manufacturing precision of the predicted values deteriorates and texture blurring occurs in sharp texture regions
Solution Approach 1:
The picture frame is divided into multiple blocks, and each block is further divided into sub-blocks for independent prediction mode determination. This segmentation allows different prediction modes to be applied to different regions, improving predicted value accuracy without significantly increasing overall system complexity.
Solution Approach 2:
Different predicted value derivation modes are selectively applied to different blocks based on their local characteristics such as motion information and texture properties. Blocks with sharp textures use one mode while blocks with smooth regions use another mode, optimizing accuracy locally while maintaining reasonable overall complexity.
2Manufacturing precision
If different predicted value derivation modes are applied to each block based on motion information and position, then the predicted value accuracy is improved and texture blurring is reduced, but the device complexity increases
Solution Approach 1:
The prediction mode for each block is dynamically determined based on motion information and position characteristics rather than being fixed. This allows the system to adaptively select appropriate modes, improving accuracy while the dynamic nature prevents excessive complexity by avoiding rigid multi-mode structures.
Solution Approach 2:
The system changes parameters such as motion vector precision and prediction mode selection based on block characteristics. By adjusting these parameters adaptively, the system achieves high predicted value accuracy in critical regions while maintaining lower complexity in less critical regions.
3Manufacturing precision
If motion information of adjacent blocks is considered in determining predicted values, then the predicted value accuracy is improved, but the processing time and productivity are reduced
Solution Approach 1:
Motion information of adjacent blocks is obtained and prepared in advance before performing the actual prediction calculation. This preliminary action allows the system to efficiently utilize pre-computed data, improving predicted value accuracy without significantly increasing processing time during the main encoding/decoding operation.
Data Source
AI summary
Provided are an inter prediction method and a terminal. The inter prediction method includes: obtaining target information, where the target information includes a predicted value derivation mode corresponding to a target picture frame and/or a predicted value derivation mode corresponding to each first block in the target picture frame; and performing inter prediction on each first block based on the target information, where the target picture frame is a to-be-encoded picture frame, and the first block is a to-be-encoded block; or the target picture frame is a to-be-decoded picture frame, and the first block is a to-be-decoded block.