Image Decoder Multiple SAO Offset Application
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing demand for high-resolution and high-quality images in various fields poses a challenge due to the diversified capabilities of terminal devices and varying network environments, necessitating an improvement in image decoding methods to enhance subjective and objective image quality.
Innovation Solution
The implementation of a method and apparatus that utilize multiple Sample Adaptive Offsets (SAOs) for image decoding, including edge and band offset-typed SAOs, to improve the quality of reconstructed images by applying these offsets to current blocks during the decoding process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If a single SAO procedure is applied to a reconstructed block, then the decoding complexity is low, but the image quality improvement is insufficient for blocks with multiple edges or wide sample value distributions
Solution Approach 1:
The patent divides the reconstructed block into multiple regions based on edge directions and applies different SAO offset procedures to each region. Specifically, it identifies multiple edges within a block and performs separate SAO operations for each edge direction, allowing precise quality improvement for complex blocks while maintaining efficiency through selective application.
Solution Approach 2:
The patent dynamically selects and applies multiple SAO offset procedures based on the characteristics of the reconstructed block. It determines whether to apply one or two SAO procedures by analyzing edge information and sample value distributions, adapting the decoding complexity to match the actual block complexity rather than applying a fixed procedure to all blocks.
2Manufacturing precision
If multiple SAO procedures are applied to improve image quality, then the subjective and objective image quality improves, but the computational complexity and processing time increase
Solution Approach 1:
The patent performs preliminary analysis of the reconstructed block to identify edge directions and determine which SAO procedures are needed before applying the offsets. By pre-processing the block to detect edges and classify regions, it avoids unnecessary SAO operations on simple blocks while preparing the appropriate procedures in advance for complex blocks, reducing overall processing time.
Solution Approach 2:
The patent applies different SAO offset procedures to different local regions of the reconstructed block based on their specific characteristics. Instead of applying a uniform procedure to the entire block, it identifies regions with different edge orientations and sample value distributions, and applies tailored SAO operations only where needed, improving quality efficiently.
Data Source
AI summary
An image decoding method performed by a decoding apparatus, according to the present invention, comprises the steps of: acquiring information on a first SAO and information on a second SAO of a current block through a bit stream; generating a reconstructed block of the current block; and performing an SAO procedure for the reconstructed block on the basis of the first SAO and the second SAO, wherein the information on the first SAO comprises first offset type information applied to the first SAO and first offset value information for n offset values, and the information on the second SAO comprises second offset type information applied to the second SAO and second offset value information for m offset values. According to the present invention, an efficient SAO procedure can be performed on the basis of a plurality of SAOs, and the video quality of images can be further improved.