NSST Image Coding for Lower Bit Rate in High-Resolution Video
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing demand for high-resolution and high-quality images leads to a significant increase in transmission and storage costs due to the large amount of information required, necessitating a more efficient image compression technique.
Innovation Solution
A video decoding method and apparatus utilizing Non-Separable Secondary Transform (NSST) to derive a range of NSST indices based on specific conditions of a target block, determining whether to encode the NSST index based on transform coefficients, and reducing bit amounts for improved coding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high-resolution and high-quality images are transmitted or stored, then image quality is improved, but transmission cost and storage cost increase
Solution Approach 1:
The patent extracts and transmits only the essential visual information by applying transform coding to convert image data into transform coefficients, where only significant coefficients are retained and transmitted. This separates the essential image information from redundant data, achieving high-quality reconstruction with reduced bit rate.
Solution Approach 2:
The patent changes the representation parameters of image data by applying different transform types (DST, DCT, DST-II) based on prediction mode and block characteristics. This parameter adaptation optimizes the compression efficiency for different image content types, maintaining quality while reducing the quantity of data to be transmitted.
2Quantity of substance
If conventional transform coding is used for image compression, then transmission cost is reduced, but coding efficiency is insufficient for high-resolution images
Solution Approach 1:
The patent introduces dynamic selection of transform types based on prediction mode (intra/inter), block size, and gradient characteristics. The transform coding parameters are adaptively changed according to local image features, making the compression system dynamically optimized for different content types rather than using a fixed transform approach.
Solution Approach 2:
The patent changes transform parameters (type selection, block size, application position) based on image content analysis including gradient calculation and prediction mode detection. This parameter adaptation enables the system to achieve better compression efficiency for high-resolution images by matching the transform characteristics to the local image structure.
3Measurement precision
If NSST index is always encoded for every block, then transform precision is improved, but bit rate increases
Solution Approach 1:
The patent applies NSST index encoding selectively only to specific blocks that meet certain criteria (e.g., gradient magnitude threshold, specific prediction modes, block size conditions) rather than uniformly to all blocks. This local application ensures transform precision is improved only where beneficial, while avoiding unnecessary bit rate increase in blocks where NSST would not provide significant improvement.
Solution Approach 2:
The patent applies NSST encoding partially rather than completely across all blocks. By using conditional encoding based on gradient analysis and block characteristics, the system applies the more precise transform only when needed, achieving a balance between transform precision and bit rate efficiency.
Data Source
AI summary
An image decoding method performed by means of a decoding device according to the present disclosure comprises the steps of: deriving transform coefficients of a target block from a bitstream; deriving a non-separable secondary transform (NSST) index with respect to the target block; performing inverse transform with respect to the transform coefficients of the target block on the basis of the NSST index and thus deriving residual samples of the target block; and generating a reconstructed picture on the basis of the residual samples.