Intra Prediction Circuit Size Reduction via Pseudo Local Decode Image
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image encoding devices face challenges in reducing circuit size while maintaining predictive image quality due to the high calculation costs and numerous prediction modes in intra prediction techniques, often resulting in lower prediction accuracy when allocating intra prediction mode selection and creation to different pipeline stages.
Innovation Solution
The solution involves creating pseudo local decode image data by encoding and decoding original image data instead of local decode image data, using Hadamard transformation and its inverse, which reduces arithmetic processing load and eliminates the need for parallel circuits for various intra prediction modes, allowing the intra prediction mode selection to be performed without directly using original image data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If parallel circuits are arranged for all intra prediction modes to create local decode images, then prediction accuracy is improved, but circuit size increases
Solution Approach 1:
The patent segments the intra prediction process into two separate pipeline stages: mode selection stage and predictive image creation stage. The mode selection stage uses simplified pseudo local decode image data to determine the optimal prediction mode, while the predictive image creation stage uses the full local decode image data. This segmentation allows accurate mode selection without requiring parallel circuits for all modes, thus reducing circuit size while maintaining prediction accuracy.
Solution Approach 2:
The patent changes the parameter of image data quality by introducing pseudo local decode image data with lower arithmetic processing load for mode selection purposes. This parameter change allows the system to use simplified data for mode determination while preserving the option to use higher quality data for actual predictive image creation, thereby resolving the contradiction between accuracy and circuit complexity.
2Loss of time
If intra prediction mode selection and predictive image creation are allocated to different pipeline stages, then processing time is reduced, but prediction accuracy deteriorates
Solution Approach 1:
The patent introduces pseudo local decode image data as an intermediary element between the mode selection stage and the predictive image creation stage. This intermediary allows the mode selection to be performed with simplified data in the first pipeline stage, while the final predictive image is created using the full local decode image data in the second stage, thus maintaining accuracy while reducing processing time through parallelization.
3Device complexity
If the number of prediction modes is decreased to suppress circuit size growth, then device complexity is reduced, but prediction accuracy is compromised
Solution Approach 1:
The patent performs preliminary action by using pseudo local decode image data to select the optimal intra prediction mode before creating the actual predictive image. This preliminary mode selection based on simplified data allows the system to maintain high prediction accuracy by choosing the best mode, while avoiding the need to implement parallel circuits for all possible modes, thus reducing device complexity.
Data Source
AI summary
The invention aims at reducing the circuit size while maintaining the creation of a predictive image suitable for the intra prediction based on a pipeline processing. The encoding device is operated to perform intra prediction in blocks of pixels resulting from division of a frame of a moving image has a pseudo-local-decode-image-creation module for simply encoding an original image, and an intra prediction module for using the created pseudo local decode image to select an intra prediction mode. In the intra prediction based on pipeline processing, the pipeline stage for selecting an intra prediction mode and the pipeline stage for creating predictive image data according to the selected intra prediction mode are separated.