Video Block Vector Matching for Efficient Intra Block Copy Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies, such as HEVC and VVC, face challenges in improving coding efficiency for video processing, particularly in intra block copy (IBC) mode, which is crucial for screen content materials.
Innovation Solution
The proposed method involves determining block vector (BV) candidates using template matching processes modified by coding factors, applying template matching costs to improve coding effectiveness and efficiency, and performing conversions based on these processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional template matching is used for block vector prediction, then coding complexity is reduced, but coding efficiency deteriorates due to inability to utilize unreconstructed samples
Solution Approach 1:
The patent performs preliminary reconstruction of samples within the current block before template matching. By reconstructing samples using already decoded neighboring blocks and previously processed portions of the current block, the method prepares reference data in advance that would traditionally be unavailable, enabling more accurate block vector prediction without adding significant computational overhead during the matching phase
Solution Approach 2:
The patent dynamically adjusts the template matching process by modifying cost functions to account for the partial reconstruction status of reference blocks. Instead of requiring fully reconstructed reference blocks, the system adapts the matching criteria to work with partially reconstructed samples, allowing flexible utilization of available data while maintaining prediction accuracy
2Measurement precision
If fully reconstructed reference blocks are required for template matching, then prediction accuracy is improved, but memory consumption and processing time increase
Solution Approach 1:
The patent applies partial action by requiring only sufficient reconstruction of reference blocks rather than complete reconstruction. The template matching process is adapted to work with partially reconstructed reference blocks, using available samples and estimating or interpolating missing values, thereby achieving acceptable prediction accuracy without the full memory and processing costs of complete reconstruction
Solution Approach 2:
The patent treats unreconstructed or partially reconstructed sample data as usable temporary resources during the encoding process. Rather than requiring persistent storage of fully reconstructed reference blocks, the system leverages transiently available sample data from the current block's processing context, effectively using disposable computational resources to achieve prediction goals
Data Source
AI summary
Embodiments of the present disclosure provide a solution for video processing. In a method, for a conversion between a current video block of a video and a bitstream of the video, at least one block vector (BV) candidate of the current video block is determined. The at least one BV candidate is associated with at least one reference block of the current video block. The at least one reference block comprises at least one reconstructed sample and at least one unreconstructed sample in a current picture. At least one template matching process is applied to the at least one BV candidate based on at least one template matching cost of the at least one BV candidate. The at least one template matching cost is modified based on at least one factor of at least one coding process. The conversion is performed based on the at least one template matching process.


