CNT Video Coding Non-Square Blocks RD Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video compression techniques face challenges in achieving high-quality prediction signals due to the inability to utilize statistical dependencies and the need for prediction values based only on initial data, leading to inefficient processing of high spatial resolution and high frame rate video contents.
Innovation Solution
The method employs Conditionally Non-linear Transform (CNT) coding on non-square blocks, utilizing previously reconstructed signals for prediction and transform coding, and optimizing block partitions based on Rate Distortion (RD) cost to enhance compression efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If prediction coding is used to obtain prediction error samples, then processing can be performed, but statistical dependency cannot be utilized and prediction quality is limited
Solution Approach 1:
The patent applies transform coding as a preliminary action before prediction coding. By first transforming the current block and then performing prediction on the transform coefficients rather than raw pixel data, the system enables utilization of statistical dependencies in the transformed domain while maintaining processing feasibility. This preliminary transform step allows subsequent prediction operations to achieve higher quality results.
2Measurement precision
If transformation coding is used to acquire prediction values, then processing can be performed, but only the first available data can be utilized and prediction quality remains limited
Solution Approach 1:
The patent transitions from spatial domain prediction to frequency domain prediction by applying transform coding. This dimensionality change from spatial to frequency domain allows the prediction process to access and utilize statistical dependencies across multiple data points that are not apparent in the spatial domain, thereby improving prediction value quality while utilizing more available data.
3Productivity
If conventional compression coding is used, then data can be compressed, but compression efficiency is insufficient for high spatial resolution and high frame rate video contents
Solution Approach 1:
The patent replaces conventional mechanical prediction coding with a hybrid approach that substitutes transform coding for the initial processing step. This substitution enables the system to handle high spatial resolution and high frame rate video contents more efficiently by exploiting statistical dependencies in the transformed domain, thereby improving compression efficiency while maintaining information quality.
Data Source
AI summary
The present invention provides a method for encoding a video signal comprising the steps of: performing block division with respect to a target unit of the video signal, wherein the block division is performed on the basis of square division and/or non-square division; determining the optimal block partition on the basis of a rate distortion cost, when the target unit is divided into a plurality of non-square blocks; and performing CNT prediction coding on the non-square blocks resulting from the optimal block partition, wherein the CNT prediction coding involves performing prediction using all previously decoded pixel values.