Transformation-Based Image Coding with Region-Based MTS Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing demand for high-resolution and high-quality images/videos, particularly in immersive media formats like VR and AR, necessitates a more efficient image/video compression technique to reduce transmission and storage costs.
Innovation Solution
An image coding method and apparatus that utilizes a multiple transform selection (MTS) index to determine the presence of significant coefficients in specific regions of a block, applying a transform kernel to derive residual samples, and encoding/decoding processes to enhance coding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high-resolution and high-quality image/video data is transmitted or stored, then image quality and resolution are improved, but transmission cost and storage cost increase
Solution Approach 1:
The patent divides the transform coefficient block into multiple regions (first region containing significant coefficients and second region with less significant coefficients). By segmenting the block, the encoder can apply different coding strategies to different regions, focusing bit allocation on regions with significant coefficients while reducing or skipping coding for regions with less significant coefficients, thus achieving compression without substantial quality loss
Solution Approach 2:
The patent applies local quality by differentiating the coding treatment based on the importance of coefficient regions. The first region with significant coefficients receives full coding attention, while the second region with less significant coefficients receives reduced attention or skip coding. This local differentiation optimizes the balance between compression ratio and perceived image quality
2Productivity
If conventional image compression techniques are used, then transmission and storage costs are reduced, but coding efficiency for high-resolution and immersive media is insufficient
Solution Approach 1:
The patent introduces dynamic adaptability through multiple transform selection (MTS), where the encoder can dynamically select from multiple transform types (e.g., DCT, DST, ADCT) based on the specific characteristics of the image or video block. This dynamic selection allows the system to adapt to various content types and resolution requirements, improving coding efficiency for high-resolution and immersive media while maintaining versatility
Solution Approach 2:
The patent utilizes parameter changes by varying transform types and parameters based on block characteristics. The encoder adjusts transform selection parameters (such as choosing different transform kernels or coefficients) according to the content properties, enabling optimized compression for different media types and resolutions
3Productivity
If transform coefficients are coded without region-based optimization, then coding simplicity is maintained, but coding efficiency is reduced
Solution Approach 1:
The patent segments the transform coefficient block into multiple regions based on scan order and coefficient significance. This segmentation enables region-based optimization where different coding methods can be applied to different regions, improving overall coding efficiency while managing complexity through systematic division of the coding task
Solution Approach 2:
The patent performs preliminary actions by pre-determining the scan order and identifying significant coefficients before the actual coding process. This preliminary analysis allows the encoder to prepare region definitions and coding strategies in advance, reducing the complexity of real-time decision-making while optimizing coding efficiency
Data Source
AI summary
An image decoding method according to the present document comprises the steps of: determining whether or not an effective coefficient is present in a second area not including an upper-left first area of a current block; parsing an MTS index from a bitstream on the basis that the effective coefficient is not present in the second area; and deriving residual samples for the current block by applying a transform kernel, which is derived on the basis of the MTS index, to transform coefficients in the first area, wherein the MTS index can be parsed on the basis that the effective coefficient, which is present in a scan sub-block which is scanned for the effective coefficient, is not present in the second area.