Block Vector Search for Intra-Block Copy Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding techniques face inefficiencies in predicting block vectors for intra-block copy and inter-prediction, leading to increased bitrate and computational complexity in video encoding and decoding processes.
Innovation Solution
The proposed techniques involve deriving a base block vector and refining it using decoder-side motion vector refinement, signaling a search process to improve the efficiency of vector prediction, and using merge mode vector prediction to reduce bitrate by selecting the best search process for vector differences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of substance
If merge mode vector prediction is used to predict block vectors, then bitrate is reduced, but prediction accuracy and efficiency deteriorate
Solution Approach 1:
The patent introduces an intermediary refinement process (DMVR) that operates between the initial merge mode prediction and the final prediction block generation. This intermediary step uses template matching to refine the motion vector, achieving both bitrate efficiency (by starting from merge mode) and prediction accuracy (through refinement).
Solution Approach 2:
The patent performs preliminary action by selecting merge candidates and generating initial vector predictions before the final refinement step. This preliminary prediction provides a good starting point that reduces the search space for subsequent refinement, improving overall efficiency.
2Measurement precision
If decoder-side motion vector refinement with template matching is performed, then prediction accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent applies partial action by performing template matching refinement only on selected merge candidates rather than all possible candidates. This selective refinement approach maintains prediction accuracy for important candidates while reducing the overall computational burden.
Solution Approach 2:
The patent segments the refinement process into distinct stages: merge candidate selection, template matching refinement, and final prediction block generation. This segmentation allows each stage to be optimized independently, managing computational complexity while maintaining accuracy.
3Productivity
If multiple search processes are evaluated and signaled, then vector prediction efficiency is improved, but signaling overhead and complexity increase
Solution Approach 1:
The patent introduces dynamic selection of search processes based on block characteristics and prediction conditions. The system adaptively chooses between different search strategies (e.g., full search, reduced search, template matching) rather than using a fixed approach, improving efficiency while managing complexity through conditional logic.
Data Source
AI summary
An example device for decoding video data includes a memory configured to store video data; and a processing system comprising one or more processors implemented in circuitry, the processing system being configured to: determine that vector information for a current block of video data is to be coded using a merge with vector difference mode; determine a search process to be used to determine a vector difference for the vector information; select a merge candidate from a merge candidate list to determine a vector predictor for the vector information; perform the search process to determine the vector difference; apply the vector difference to the vector predictor to form a final vector, and form a prediction block for the current block using the final vector.