Co-located Block Position Adjustment for Video Encoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional inter prediction coding methods in moving picture coding do not adequately leverage co-located blocks across different views to enhance coding efficiency.
Innovation Solution
A moving picture coding method that determines whether a current picture and a co-located block are in the same view, adjusts the position of the co-located block using a disparity vector, and adds the adjusted motion vector to the list, allowing for more accurate prediction and improved coding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional inter prediction coding is used without adjusting co-located block positions across different views, then the coding process is simple, but coding efficiency is insufficient
Solution Approach 1:
The patent applies preliminary action by determining view relationships and adjusting co-located block positions before performing motion compensation. The coding device预先 determines whether the current picture and co-located block are in the same view, and if in different views, adjusts the co-located block position using disparity vectors before deriving motion vectors. This preliminary positioning ensures more accurate prediction while maintaining a structured, manageable process flow.
2Measurement precision
If co-located blocks from different views are used without position adjustment, then the reference picture list is simple to construct, but prediction accuracy is reduced
Solution Approach 1:
The patent applies local quality by treating co-located blocks in different views differently from those in the same view. When the current picture and co-located block are in different views, the patent adjusts the co-located block position using view-specific disparity vectors. This localized adjustment ensures that each co-located block is positioned appropriately according to its specific view relationship, improving prediction accuracy without uniformly complicating the entire reference picture list construction process.
3Measurement precision
If disparity vector calculation and co-located block position adjustment are performed, then motion prediction accuracy improves, but computational complexity increases
Solution Approach 1:
The patent applies partial action by selectively adjusting co-located block positions only when necessary - specifically when the current picture and co-located block are determined to be in different views. The view determination step allows the system to skip adjustment operations when views are the same, reducing unnecessary computational overhead. This selective approach applies disparity vector calculations only where needed, balancing accuracy improvement with computational efficiency.
Data Source
AI summary
A moving picture coding method including: determining whether or not (a) a picture including a co-located block and (b) a current picture to be coded are included in a same view, the co-located block being a block that is included in a picture different from the current picture and is at a position corresponding to a position of a current block to be coded included in the current picture; adjusting the position of the co-located block when the picture including the co-located block and the current picture are included in different views; and adding to the list an entry including a motion vector derived from the co-located block, wherein the adjusting includes: obtaining a disparity vector between the view including the picture including the co-located block and the view including the current picture; and adjusting the position of the co-located block by the obtained disparity vector.