Encoder Partition Prediction Gradient Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image encoding technologies face challenges in efficiently processing image blocks, particularly when splitting them into non-rectangular partitions, which complicates prediction processing and increases processing amounts.
Innovation Solution
An encoder that includes circuitry and memory, which splits an image block into partitions, determines whether each partition is non-rectangular, and adjusts the prediction image generation process accordingly, either using gradient images for rectangular partitions or omitting them for non-rectangular partitions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the encoder processes non-rectangular partitions using gradient images, then the prediction accuracy improves, but the processing amount increases
Solution Approach 1:
The patent applies different prediction processing methods to different partition types: rectangular partitions use gradient image processing for high accuracy, while non-rectangular partitions use simplified processing. This local differentiation optimizes the balance between prediction accuracy and processing complexity for each specific case.
Solution Approach 2:
The patent segments the partition processing into two distinct paths: one for rectangular partitions that includes gradient image generation and application, and another for non-rectangular partitions that uses simplified prediction. This segmentation allows the system to avoid unnecessary complex processing for partition types where it provides minimal benefit.
2Productivity
If the encoder uses gradient images for all partitions, then the encoding efficiency improves, but the processing complexity increases
Solution Approach 1:
The patent implements conditional processing where gradient images are generated and applied only to rectangular partitions, while non-rectangular partitions receive simplified prediction processing. This selective application maintains encoding efficiency for suitable partition types while avoiding unnecessary complexity for others.
3Device complexity
If the encoder processes non-rectangular partitions with simplified methods, then the processing amount reduces, but the prediction accuracy decreases
Solution Approach 1:
The patent divides partition processing into separate handling paths: rectangular partitions receive full gradient-based prediction processing for high accuracy, while non-rectangular partitions use alternative simplified methods. This segmentation ensures that complex processing is applied only where it provides significant benefit.
Solution Approach 2:
The patent changes the prediction processing parameters based on partition geometry: for non-rectangular partitions, it adjusts the prediction method to account for the irregular shape, using alternative approaches that maintain acceptable accuracy while reducing processing complexity compared to applying gradient methods to all partition types.
Data Source
AI summary
Provided is an encoder which includes circuitry and memory. Using the memory, the circuitry splits an image block into a plurality of partitions, obtains a prediction image for a partition, and encodes the image block using the prediction image. When the partition is not a non-rectangular partition, the circuitry obtains (i) a first prediction image for the partition, (ii) a gradient image for the first prediction image, and (iii) a second prediction image as the prediction image using the first prediction image and the gradient image. When the partition is a non-rectangular partition, the circuitry obtains the first prediction image as the prediction image without using the gradient image.