Adaptive DIMD Filtering for Accurate Intra Mode Derivation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies face challenges in efficiently determining intra prediction modes for video blocks, particularly in decoder-side intra mode derivation, leading to suboptimal compression and quality degradation.
Innovation Solution
The method involves processing a bitstream with syntax elements indicating filter types for left and top templates adjacent to a current block, applying decoder-side intra mode derivation (DIMD) mode based on these filters to determine intra prediction modes, using Sobel filters with specific horizontal and vertical components to enhance prediction accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If decoder-side intra mode derivation (DIMD) is applied to determine intra prediction modes, then prediction accuracy is improved, but computational complexity increases
Solution Approach 1:
The template is divided into multiple regions (e.g., top template and left template) that can be processed independently. Different filter types can be applied to different regions based on their characteristics, reducing the overall computational complexity while maintaining prediction accuracy.
Solution Approach 2:
Instead of applying complex filtering to the entire template, the method applies filters selectively to specific regions or samples within the template. This partial action approach reduces computational load while still capturing the essential gradient information needed for accurate prediction.
2Measurement precision
If multiple filter types are applied to templates, then determination of intra prediction modes is improved, but processing time increases
Solution Approach 1:
The filter type selection is made dynamically based on the characteristics of each template region rather than using a fixed filter for all regions. This allows the system to adaptively choose the most appropriate filter for each region, improving accuracy while avoiding the need to apply all filter types everywhere.
Solution Approach 2:
Different filter parameters (kernel size, orientation) are applied to different template regions based on local characteristics. This parameter adaptation allows accurate prediction mode determination without uniformly increasing processing time across the entire template.
3Measurement precision
If Sobel filters with specific horizontal and vertical components are used, then prediction accuracy is enhanced, but device complexity increases
Solution Approach 1:
The Sobel filter is applied with different orientations (horizontal for top template, vertical for left template) matched to the local geometry of each template region. This local adaptation of filter quality enhances prediction accuracy while keeping the filter implementation straightforward by using standard Sobel kernels in appropriate orientations.
Data Source
AI summary
Methods and apparatuses for video decoding and video encoding and a method of processing visual media data are disclosed. The apparatus for video decoding includes processing circuitry that receives coded information indicating that a current block in a current picture is coded with a decoder-side intra mode derivation (DIMD) mode. A template of the current block includes reconstructed samples in the current picture and is adjacent to the current block. The template includes one of a left template and a top template. The processing circuitry determines a filter type from a plurality of filter types associated with the one of the left template and the top template, applies the DIMD mode to the template based on the determined filter type to determine one or more intra prediction modes for the current block, and reconstructs the current block according to the one or more intra prediction modes.