Image Decoding Pixel Allocation Across Slice Boundaries
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In image coding and decoding processes, particularly in H.264, the division of pictures into slices leads to limitations in intra-frame prediction, resulting in decreased image quality and coding efficiency due to restricted information use and forced changes in motion vectors at slice boundaries, which negatively impact coding efficiency.
Innovation Solution
An image decoding method that allocates the value of an inside pixel to an outside pixel across slice boundaries in a reference picture, using stored identification information to determine the range of slices and apply pixel stretching, thereby reconstructing blocks and improving prediction accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the picture is divided into slices for parallel processing, then processing efficiency is improved, but image quality and coding efficiency deteriorate due to restricted information use and forced motion vector changes at slice boundaries
Solution Approach 1:
The patent applies local quality by treating different regions (slices) differently in terms of prediction information availability. At slice boundaries, the system uses pixel stretching to generate prediction images from available reference data, while maintaining normal prediction processes within slices. This allows parallel processing of slices while preserving image quality through localized adaptation at boundaries.
2Measurement precision
If pixel stretching is applied to generate prediction images at slice boundaries, then prediction accuracy is improved, but device complexity increases due to additional processing steps
Solution Approach 1:
The patent implements preliminary action by pre-defining the pixel stretching process as a standard procedure for handling slice boundary cases. The system automatically identifies when pixel stretching is needed (at slice boundaries) and applies the appropriate prediction image generation method without requiring complex real-time decision-making, thus managing processing complexity while maintaining high prediction accuracy.
Data Source
AI summary
An image coding method is provided for coding, on a per-block basis, pictures each including slices. The method involves estimating a current motion vector which is a motion vector of a current block to be coded and specifies a reference block included in a reference picture; generating a prediction image block by allocating a value of an inside pixel to an outside pixel, the inside pixel being a pixel located inside an associated slice, the outside pixel being a pixel located outside the associated slice and included in the reference block specified by the current motion vector, and the associated slice being a slice included in the reference picture and corresponding to a current slice to be coded which includes the current block; subtracting the prediction image block from the current block to generate a difference image block; and coding the current motion vector and the difference image block.