History-Based Motion Vector Prediction Merge Candidate List
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video compression technologies face challenges in efficiently processing next-generation video contents with high spatial resolution, high frame rate, and high dimensionality, leading to increased memory storage and processing demands.
Innovation Solution
The proposed solution involves configuring a merge candidate list based on neighboring blocks, adding history-based merge candidates and zero motion vectors to the list when necessary, and updating the list based on motion information to enhance prediction efficiency and reduce redundancy checks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a history-based motion vector prediction table (HMVP table) is maintained and used for inter prediction, then prediction accuracy is improved, but memory storage requirements and processing complexity increase
Solution Approach 1:
The patent extracts only the essential motion information (motion vectors and reference picture indices) from the full motion data structure and stores only these critical elements in the HMVP table. This selective extraction reduces the memory footprint and processing overhead while retaining the most important prediction capabilities.
Solution Approach 2:
The patent applies different data structures and storage strategies to different types of motion information. Frequently accessed motion vectors are stored in a compact format in the HMVP table, while less critical data is handled differently. This local optimization reduces overall complexity while maintaining prediction accuracy where it matters most.
2Measurement precision
If the HMVP table size is increased to store more motion vector candidates, then prediction performance is improved, but memory storage requirements increase
Solution Approach 1:
The patent implements a limited-size HMVP table that stores only a subset of available motion vector candidates (typically 1-4 entries). This partial approach provides sufficient prediction performance for most cases while avoiding the memory overhead of storing all possible candidates. The table is selectively populated based on prediction needs rather than maintaining a complete list.
Solution Approach 2:
The patent dynamically adjusts the HMVP table size and content based on prediction performance requirements and available memory resources. The table can be configured to store different numbers of candidates depending on the specific application needs, allowing optimization between prediction accuracy and memory usage.
3Reliability
If redundancy checks are performed when adding HMVP candidates to the merge candidate list, then prediction reliability is improved, but processing time increases
Solution Approach 1:
The patent performs preliminary validation of motion information before adding candidates to the HMVP table and merge candidate list. By checking for duplicate motion vectors and reference picture indices in advance, the system avoids the need for more extensive redundancy checks later, reducing overall processing time while maintaining reliability.
Solution Approach 2:
The patent implements optimized redundancy checks that skip unnecessary verification steps when motion information is clearly unique. When a candidate motion vector differs significantly from existing entries, the system rapidly adds it without performing full redundancy validation, thus reducing processing time for obvious cases while maintaining reliability for borderline cases.
Data Source
AI summary
The disclosure discloses a method for processing video signals and an apparatus therefor. Specifically, the method of processing video signals based on history based motion vector prediction, comprising: configuring a merge candidate list based on a neighboring block to a current block; adding a history based merge candidate of the current block to the merge candidate list when a number of merge candidates included in the merge candidate list is smaller than a first predetermined number; adding a zero motion vector to the merge candidate list when the number of merge candidates included in the merge candidate list is smaller than a second predetermined number; obtaining a merge index indicating a merge candidate used for inter prediction of the current block in the merge candidate list; generating a prediction sample of the current block based on motion information of the merge candidate indicated by the merge index; and updating a history based merge candidate list based on the motion information.