Picture Decoding Block Partitioner Boundary Constraint
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing techniques for picture coding and decoding face a decrease in coding efficiency when blocks are not partitioned into appropriate sizes and shapes, particularly at the frame end where blocks contain pixels beyond the picture boundary.
Innovation Solution
A picture decoding device is developed, equipped with a decoder to decode block partitioning information and a block partitioner that recursively partitions the picture into rectangles of a predetermined size. The block partitioner includes a quad splitter and a binary/ternary splitter, which restrict partitioning when it would result in blocks beyond the picture boundary, ensuring appropriate block sizes and shapes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If blocks are partitioned recursively without restrictions, then the picture can be divided into fine-grained blocks for better prediction accuracy, but blocks at the frame end will extend beyond the picture boundary causing inappropriate block sizes and shapes that decrease coding efficiency
Solution Approach 1:
The patent applies different partitioning rules to different regions of the picture. Blocks at the frame end (near picture boundaries) are subject to restricted partitioning to prevent extending beyond boundaries, while blocks in the interior region can be partitioned more freely. This local differentiation ensures that each region is processed with appropriate block sizes and shapes, maintaining prediction accuracy without compromising coding efficiency at boundaries.
2Ease of operation
If blocks at the frame end are allowed to extend beyond the picture boundary, then recursive partitioning can be completed uniformly, but the resulting blocks have inappropriate sizes and shapes that decrease coding efficiency
Solution Approach 1:
The patent segments the picture processing into two distinct parts: interior blocks and boundary blocks. By identifying blocks near the picture boundary and applying restricted partitioning rules specifically to these boundary blocks, the system maintains uniform processing for interior blocks while preventing inappropriate extensions at the frame end. This segmentation approach preserves coding efficiency by ensuring all blocks have appropriate sizes and shapes for their respective locations.
Data Source
AI summary
A block partitioner includes a quad splitter structured to partition a target block obtained by recursive partitioning in half in both a horizontal direction and a vertical direction to generate four blocks, and a binary/ternary splitter structured to partition the target block obtained by recursive partitioning into two or three in the horizontal direction or the vertical direction to generate two or three blocks, and the binary/ternary splitter disallows partitioning of the target block in the horizontal direction when partitioning of the target block in the horizontal direction causes the target block obtained by partitioning to be located beyond a right side of a picture boundary, and disallows partitioning of the target block in the vertical direction when partitioning of the target block in the vertical direction causes the target block obtained by partitioning to be located beyond a lower side of the picture boundary.