Intra Block Copy Merging Syntax for Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding standards lack flexibility and efficiency in handling different types of content, as the maximum number of intra block copy merging candidates is not separately signaled for intra block copy mode, leading to suboptimal performance for natural and screen content.
Innovation Solution
Introducing a separate syntax element to specify the maximum number of intra block copy merging candidates, allowing for more flexible and efficient coding by enabling different configurations based on content type, such as using fewer candidates for natural content and more for screen content.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a single syntax element specifies the maximum number of merging candidates for all prediction modes, then the standard structure is simple and unified, but the coding efficiency is suboptimal for different content types
Solution Approach 1:
The patent segments the merging candidate configuration by introducing a separate syntax element (max_cb_merge_cand) specifically for intra block copy mode, distinct from the general merge mode syntax element (max_merge_cand). This segmentation allows independent optimization of candidate numbers for different prediction modes and content types without affecting other modes, resolving the contradiction between adaptability and structural simplicity.
2Productivity
If more merging candidates are used for intra block copy mode, then the coding efficiency improves for screen content, but the signaling overhead and processing complexity increase
Solution Approach 1:
The patent applies local quality by allowing different maximum numbers of merging candidates for different prediction modes. Specifically, intra block copy mode can be configured with a different max_cb_merge_cand value than the general max_merge_cand used for other modes. This enables optimization tailored to local content characteristics (screen content vs. natural content) without unnecessarily increasing the candidate number for all modes, thus balancing coding efficiency with signaling overhead.
Data Source
AI summary
A method of encoding and decoding video data, including coding a first syntax element that specifies a value used to derive a maximum number of intra block copy merging candidates, deriving the maximum number of intra block copy merging candidates based on the value of the first sytnax element, and coding a first block of video data using intra block copy mode according to the maximum number of intra block copy merging candidates.