Video Coding Apparatus Prediction Mode Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding standards, such as HEVC and its extension SCC, face inefficiencies in determining optimal prediction modes and scaling coefficients, leading to suboptimal coding efficiency due to the lack of systematic methods for evaluating and selecting these parameters.
Innovation Solution
A video coding apparatus and method that calculates evaluation values for various prediction modes and scaling coefficients based on luminance and color difference prediction errors, selects the most suitable modes and coefficients, and adjusts them to minimize prediction errors, thereby improving coding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If multiple prediction modes and scaling coefficients are evaluated to improve coding efficiency, then the coding precision is improved, but the computational complexity and processing time increase
Solution Approach 1:
The patent applies preliminary action by calculating evaluation values for multiple prediction modes and scaling coefficients before final selection. The encoder computes prediction errors and evaluation metrics in advance for all candidate modes, then selects the optimal combination. This allows the system to make informed decisions about which prediction mode and scaling coefficients to use, improving coding precision while managing computational complexity through systematic pre-evaluation.
Solution Approach 2:
The patent employs parameter changes by systematically varying prediction modes and scaling coefficients to find the optimal combination. Different prediction modes (intra, inter, screen content) and scaling coefficient values are tested, and the parameters are adjusted based on evaluation values calculated from prediction errors. This parameter optimization approach improves coding precision by selecting the best combination for each coding unit.
2Productivity
If optimal prediction modes and scaling coefficients are determined through systematic evaluation, then coding efficiency is improved, but the processing time increases
Solution Approach 1:
The patent calculates evaluation values for multiple prediction modes and scaling coefficients in advance, before final encoding decisions are made. By pre-computing prediction errors and evaluation metrics for all candidate modes, the system can efficiently select the optimal combination without repeated calculations during the encoding process, thus improving coding efficiency while managing processing time.
Solution Approach 2:
The patent implements feedback by using evaluation values derived from prediction errors to guide the selection of prediction modes and scaling coefficients. The system calculates prediction errors for candidate modes, computes evaluation values based on these errors, and uses this feedback information to select the optimal mode. This feedback mechanism ensures coding efficiency by making data-driven decisions about parameter selection.
3Loss of substance
If quantization is applied to scaling coefficients to reduce data size, then the data compression is improved, but the manufacturing precision of the coefficients deteriorates
Solution Approach 1:
The patent applies parameter changes by quantizing scaling coefficients to a limited set of predefined values. Instead of using continuous scaling coefficient values, the system maps coefficients to discrete levels (e.g., 0, 0.5, 1, 1.5, 2). This quantization reduces the data size required to represent coefficients while maintaining acceptable precision through the use of strategically chosen quantization levels that preserve the essential scaling behavior.
Data Source
AI summary
An apparatus for coding video in one of a plurality of prediction modes executes a first process for calculating first evaluation values regarding each of the plurality of prediction modes, based on a plurality of luminance prediction errors, each luminance prediction error being a difference between luminance of each pixel in an input image and luminance of each pixel in a prediction image, a plurality of color difference prediction errors, each color difference prediction error being a difference between a color difference of each pixel in the input image and a color difference of each pixel in the prediction image, and a calculation result of the luminance prediction errors and the color difference prediction errors; executes a second process for selecting a predetermined number of prediction modes from the prediction modes, based on the first evaluation values.