Video Coding Prediction Mode Selection via Homogeneous Region Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding technologies, such as H.264, are inefficient in determining the optimal prediction mode due to the time-consuming process of calculating rate-distortion costs for each prediction mode, leading to redundant calculations and reduced efficiency in video coding.
Innovation Solution
An image processing method that adaptively determines the coding prediction manner of a coding unit by analyzing the prediction modes of adjacent and parent/child units, as well as preprocessing image blocks, to quickly select the optimal prediction mode and processing order, thereby reducing unnecessary calculations and improving efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If rate-distortion cost calculation is performed for each prediction mode, then the optimal prediction mode can be determined accurately, but the coding efficiency deteriorates due to time-consuming calculations
Solution Approach 1:
The patent performs preliminary classification of coding units into homogeneous and heterogeneous regions before detailed prediction mode selection. By pre-identifying homogeneous regions where all units share the same prediction mode, the system avoids redundant rate-distortion calculations for these units, thus improving coding efficiency while maintaining accurate prediction mode determination through subsequent targeted calculations
2Reliability
If prediction mode calculation is performed for all coding units, then the optimal prediction mode is selected, but redundant calculations increase processing time
Solution Approach 1:
The patent segments coding units into homogeneous regions (where all units use the same prediction mode) and heterogeneous regions (where different prediction modes are used). This segmentation allows the system to apply different processing strategies: homogeneous regions are handled efficiently with unified prediction modes, while heterogeneous regions receive detailed individual analysis, thereby reducing redundant calculations and processing time while maintaining prediction mode optimality
Solution Approach 2:
Instead of performing full prediction mode calculations for all coding units, the patent applies partial action by conducting detailed calculations only for heterogeneous regions where prediction mode variation is expected. Homogeneous regions are processed with simplified assumptions, reducing overall processing time while maintaining sufficient prediction accuracy for the majority of the video content
Data Source
AI summary
An image processing method is provided. The image processing method includes determining an associated unit set of a current coding unit in an image frame, the associated unit set comprising associated coding units, the associated coding units including a target unit in the image frame adjacent to the current coding unit, and at least one coding unit in the image frame having a parent/child relationship with the current coding unit, determining first statistical information based on a prediction mode of each associated coding unit in the associated unit set, obtaining second statistical information of preprocessing image blocks corresponding to the current coding unit, and determining a coding prediction manner of the current coding unit based on the first statistical information and the second statistical information, the coding prediction manner indicating a selected prediction mode and/or a processing order of the prediction mode.