Intra Prediction Phase Shifting for H.264/AVC Encoding Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The H.264/AVC system's intra prediction method is limited by using only adjacent pixels for prediction, leading to increased memory access and processing times, especially when predicting angles finer than 22.5 degrees, which restricts encoding efficiency improvement.
Innovation Solution
An image processing device that determines the optimal phase shift for adjacent pixels using a mode determining means, phase shift means, and shift amount determining means to generate a prediction image, allowing for improved prediction precision without increasing memory access and processing times.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If phase shifting is applied to achieve finer prediction angles (less than 22.5 degrees), then prediction precision is improved, but memory access and processing times increase
Solution Approach 1:
The patent pre-calculates and stores phase-shifted adjacent pixel values in memory before the actual prediction process. By preparing these shifted values in advance (preliminary action), the system can perform fine-angle predictions without performing complex shifting operations during real-time encoding, thus improving prediction precision while avoiding the time penalty of on-the-fly phase shifting.
Solution Approach 2:
The patent dynamically selects the appropriate phase shift amount based on the predicted prediction angle. Instead of always applying maximum phase shifting, the system adaptively adjusts the shift amount to match the required prediction precision, performing phase shifting only when and where needed. This dynamic approach maintains high prediction accuracy for fine angles while minimizing unnecessary processing for coarser angles.
2Measurement precision
If phase shifting is applied to achieve finer prediction angles (less than 22.5 degrees), then prediction precision is improved, but device complexity increases
Solution Approach 1:
The patent pre-calculates and stores phase-shifted adjacent pixel values in memory before the actual prediction process. By preparing these shifted values in advance (preliminary action), the system can perform fine-angle predictions without performing complex shifting operations during real-time encoding, thus improving prediction precision while avoiding the time penalty of on-the-fly phase shifting.
Solution Approach 2:
The patent introduces a phase shift amount determination unit that acts as an intermediary between the prediction angle calculation and the actual prediction process. This intermediary component determines the appropriate phase shift amount based on the prediction angle and controls the phase shifting operation, thereby simplifying the overall system architecture by centralizing the complexity in a dedicated module rather than distributing it throughout the prediction engine.
3Productivity
If only adjacent pixels are used for intra prediction, then memory access is reduced, but prediction precision is limited
Solution Approach 1:
The patent pre-calculates and stores phase-shifted adjacent pixel values in memory before the actual prediction process. By preparing these shifted values in advance (preliminary action), the system can perform fine-angle predictions without performing complex shifting operations during real-time encoding, thus improving prediction precision while avoiding the time penalty of on-the-fly phase shifting.
Solution Approach 2:
The patent changes the parameter of pixel position by applying phase shifting to adjacent pixels. Instead of using only the original adjacent pixel positions, the system modifies their positions through phase shifting to create virtual pixels at fractional positions. This parameter change enables the prediction to access pixel values at positions corresponding to fine prediction angles, thereby improving prediction precision while still relying on adjacent pixel data.
Data Source
AI summary
The present invention relates to an image processing device and method which enable encoding efficiency in intra prediction to be improved. In the event that the optimal intra prediction mode is mode 0, adjacent pixels to be used for prediction of the current block are pixels A0, A1, A2, and A3. According to these pixels and a 6-tap FIR filter, pixels a−0.5, a+0.5, and so on with ½ pixel precision are generated, and further, pixels a−0.75, a−0.25, a+0.25, and a+0.75 with ¼ pixel precision are generated by linear interpolation. Subsequently, the optimal shift amount is determined with a value of −0.75 through +0.75 that is phase difference between an integer pixel and generated fractional pixel precision serving as a candidate of the shift amount in the horizontal direction. The present invention may be applied to an image encoding device which performs encoding using the H.264/AVC system, for example.


