Intra Prediction Unit Gradient Adjustment for Video Encoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional moving image encoding devices struggle to achieve high-accuracy predictions when the signal value varies along the prediction direction, leading to increased prediction errors even if the object boundary direction matches the prediction direction.
Innovation Solution
The proposed solution involves an image encoding device that, during intra prediction, adds a value proportional to the change in the horizontal or vertical direction of luminance values of adjacent pixels to the luminance value of the corresponding pixel, thereby setting the result as the predicted value of the prediction image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional intra prediction methods are used, then the prediction process is simple, but prediction accuracy deteriorates when signal values vary along the prediction direction
Solution Approach 1:
The patent changes the prediction parameters by introducing gradient calculations based on luminance value changes in horizontal and vertical directions. Instead of using fixed prediction modes, the system dynamically adjusts prediction values by adding proportional changes based on adjacent pixel luminance variations, thereby improving prediction accuracy while maintaining manageable complexity
Solution Approach 2:
The patent introduces dynamic adaptation to signal variations by calculating gradient values from adjacent pixels and adjusting prediction values accordingly. The prediction process becomes dynamic rather than static, allowing the system to adapt to different signal characteristics in different regions of the image block
2Productivity
If directional prediction is used to match object boundary direction, then prediction efficiency improves for constant signal regions, but prediction error increases when signal values vary along the prediction direction
Solution Approach 1:
The patent applies local quality by treating different regions of the prediction block differently based on their local signal characteristics. By calculating gradient values from adjacent pixels and applying proportional changes locally, the system adapts to local variations in signal values rather than applying a uniform prediction approach across the entire block
Data Source
Figure 1
Figure 2
Figure 3
AI summary
When an intra prediction parameter indicates a horizontal prediction process, an intra prediction unit adds a value proportional to a change in a horizontal direction of the luminance values of pixels adjacent to the top of each block, which is a unit for prediction process of a coding block, to the luminance value of a pixel adjacent to the left of the block to set a result of the addition as a predicted value of a prediction image, whereas when the intra prediction parameter indicates a vertical prediction process, the intra prediction unit adds a value proportional to a change in a vertical direction of the luminance values of pixels adjacent to the left of the block to the luminance value of the pixel adjacent to the top of the block to set the result of the addition as a predicted value of the prediction image.