Transform Coefficient Zero-Out Coding for Residual Bit Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing demand for high-resolution and high-quality images/videos, particularly in immersive media formats, leads to higher transmission and storage costs due to increased bit amounts, necessitating a more efficient image/video compression technique.
Innovation Solution
An image coding method and apparatus that enhances efficiency by coding transform coefficients based on high-frequency zeroing, deriving context models for last significant coefficient positions, and optimizing residual coding through zero-out blocks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high-resolution and high-quality image/video data is transmitted or stored using conventional methods, then image quality is maintained, but transmission cost and storage cost increase due to increased bit amount
Solution Approach 1:
The patent extracts and removes high-frequency transform coefficients that correspond to regions outside the object of interest (background regions). By identifying the object boundary and zeroing out coefficients in the frequency domain that correspond to background areas, the method reduces the bit amount required for transmission and storage while preserving the visual quality of the foreground object.
Solution Approach 2:
The patent applies different processing strategies to different regions of the image based on their importance. High-frequency coefficients corresponding to the object region are preserved to maintain detail and quality, while high-frequency coefficients corresponding to background regions are zeroed out. This localized quality management reduces overall bit amount while maintaining perceived image quality.
2Productivity
If conventional transform coding is applied to all regions, then compression is achieved, but coding efficiency decreases for high-frequency coefficients in background regions
Solution Approach 1:
The patent performs preliminary action by zeroing out high-frequency transform coefficients corresponding to background regions before the actual encoding process. By pre-identifying the object boundary and setting irrelevant high-frequency coefficients to zero, the method eliminates unnecessary data that would otherwise consume bits, thereby improving compression efficiency without affecting perceived quality.
3Measurement precision
If all transform coefficients are coded with full precision, then coding accuracy is maintained, but residual coding efficiency decreases
Solution Approach 1:
The patent applies local quality management by maintaining full precision coding for transform coefficients corresponding to the object region while reducing or eliminating precision for coefficients corresponding to background regions. This selective precision approach maintains coding accuracy for important visual information while improving residual coding efficiency by reducing the bit cost for less important background information.
Data Source
AI summary
An image decoding method according to the present document comprises a step for deriving a residual sample, wherein the step for deriving the residual sample comprises the steps of: deriving a zero out block for a current block; deriving a context model for last significant coefficient location information on the basis of the width or height of the current block; deriving a value of a last significant coefficient location on the basis of the context model; and deriving the last significant coefficient location on the basis of a value of the last significant coefficient location information and the width or height of the zero out block.