Adaptive Inverse Transform Selection for Sparse Video Decoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing demand for high-definition and high-quality images leads to higher data volumes, increasing transmission and storage costs, and existing image compression technologies are not optimized for high-resolution and stereoscopic image content.
Innovation Solution
An image encoding/decoding method and apparatus that perform an inverse transform using linearity, determining the inverse transform method based on the number of non-zero coefficients, and applying linear inverse transforms selectively to reduce computational complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If high-resolution and high-quality image data is used, then image quality is improved, but data volume and transmission/storage costs increase
Solution Approach 1:
The patent applies parameter changes by dynamically selecting different inverse transform methods (full inverse transform vs. linearity-based inverse transform) based on the number of non-zero coefficients in the transform coefficient block. This allows the system to adapt the processing complexity and data representation based on the actual content characteristics, achieving efficient compression while maintaining image quality.
2Measurement precision
If conventional inverse transform methods are used, then decoding accuracy is maintained, but computational complexity increases
Solution Approach 1:
The patent segments the inverse transform process into two distinct methods: a full inverse transform method for blocks with many non-zero coefficients, and a simplified linearity-based inverse transform method for blocks with few non-zero coefficients. This segmentation allows the system to apply the appropriate level of computational complexity based on the actual data characteristics, reducing overall computational load while maintaining decoding accuracy where needed.
Solution Approach 2:
The patent applies partial action by using the simplified linearity-based inverse transform method only when necessary (when the number of non-zero coefficients is below a threshold), rather than applying the full inverse transform to all blocks. This partial application of the simplified method reduces computational complexity while maintaining sufficient decoding accuracy for appropriate cases.
3Productivity
If the number of non-zero coefficients is small, then compression efficiency is improved, but inverse transform accuracy may deteriorate
Solution Approach 1:
The patent applies dynamics by making the inverse transform method adaptive rather than static. The system dynamically selects between the full inverse transform method and the linearity-based method based on the actual number of non-zero coefficients in each transform coefficient block. This dynamic adaptation ensures that compression efficiency is maximized when appropriate (using linearity for sparse blocks) while maintaining inverse transform accuracy when needed (using full transform for dense blocks).
Data Source
AI summary
The present disclosure provides a video decoding method including: acquiring the number of non-zero coefficients of a dequantized block; determining an inverse transform method of the dequantized block according to the number of non-zero coefficients; and performing an inverse transform of the dequantized block according to the determined inverse transform method.