Adaptive Video Transform for Compression Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing resolution and quality of video data lead to higher data volumes, resulting in increased transfer and storage costs, necessitating more efficient video compression techniques.
Innovation Solution
A method and apparatus for video encoding and decoding that perform a first transform based on prediction and a second transform on a low-frequency domain, with the mode of the second transform determined by the first transform, to enhance compression efficiency. This includes discrete sine transform (DST) for 4×4 blocks and adaptive transform skip modes to optimize processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If video resolution and quality are increased, then video quality is improved, but data volume increases leading to higher transfer and storage costs
Solution Approach 1:
The patent divides the transform process into two stages: a first transform (e.g., DCT) applied to large blocks (8x8 or larger) and a second transform (DST) applied to smaller 4x4 sub-blocks. This segmentation allows the system to handle different frequency components differently, achieving better compression efficiency without sacrificing quality, thereby reducing the data volume needed for high-resolution video.
Solution Approach 2:
The patent applies different transform types to different regions of the frequency domain. The first transform is applied to the overall block while the second transform is applied specifically to the low-frequency domain (4x4 blocks). This local differentiation optimizes the compression for each region, improving overall compression efficiency and reducing data requirements for high-quality video.
2Productivity
If a single transform is used for the entire block, then processing is simple, but compression efficiency is insufficient for high-resolution video
Solution Approach 1:
The transform process is segmented into two distinct stages: first transform on the entire block and second transform on the low-frequency domain. This segmentation enables the system to achieve superior compression efficiency by addressing different frequency characteristics separately, while the complexity is managed through a systematic two-step approach rather than a single complex operation.
Solution Approach 2:
The patent dynamically selects transform modes (2D transform, 1D vertical transform, 1D horizontal transform, or non-transform) based on the prediction mode and block shape. This dynamic adaptation allows the system to optimize compression efficiency for different video content and conditions while managing processing complexity through intelligent mode selection rather than always performing the most complex transform.
Data Source
AI summary
The present invention relates to a technique for encoding and decoding an image, and a method for encoding the image, according to the present invention, comprises the steps of: performing a prediction with respect to a current block; and transforming a residual which is generated by the prediction, wherein in the step of transforming, a first transform is performed on the residual, and then a second transform is performed on a lower frequency area, and wherein a mode of the second transform can be decided on the basis of a mode of the first transform.