Geometric Video Block Partitioning for Motion Vector Merge Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies face challenges in efficiently encoding and decoding high-definition and ultra-high-definition video data while maintaining image quality, as the amount of data to be processed grows exponentially, necessitating improved block partitioning schemes and motion estimation methods.
Innovation Solution
Implementing geometric partitioning of coding units into triangular prediction units and performing motion vector comparison operations to enhance video coding efficiency, particularly through the use of geometric shaped prediction units and motion compensated prediction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If block-based video coding is used to compress video data, then video quality can be maintained, but the amount of data to be processed grows exponentially with high-definition and ultra-high-definition formats
Solution Approach 1:
The patent divides coding units into multiple prediction units with different shapes (triangular, rectangular, irregular) to better segment the video content. This segmentation allows for more precise representation of motion and texture patterns, maintaining video quality while reducing the overall data volume through more efficient prediction
Solution Approach 2:
The patent applies different prediction unit shapes and sizes to different regions of the video frame based on local characteristics. Motion-compensated prediction is applied selectively to regions with significant motion, while other regions use simpler prediction methods, optimizing the balance between quality and data reduction
2Productivity
If traditional block partitioning schemes are used, then encoding simplicity is maintained, but coding efficiency deteriorates for high-definition video
Solution Approach 1:
The patent introduces dynamic partitioning where the shape and size of prediction units are adaptively selected based on motion characteristics and content analysis. This dynamic approach allows the system to switch between different partitioning strategies (triangular, rectangular, irregular shapes) to optimize coding efficiency for different video regions and motion patterns
Solution Approach 2:
The patent changes multiple parameters including prediction unit shape, size, and partitioning depth to optimize coding efficiency. By varying these parameters adaptively based on video content analysis, the system achieves better compression ratios for high-definition video without requiring a complete overhaul of the encoding architecture
3Measurement precision
If motion estimation is performed to find reference blocks, then prediction accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent segments the search space for motion estimation by dividing the reference frame into multiple candidate regions based on motion patterns and content analysis. This segmentation reduces the number of blocks that need to be compared while maintaining prediction accuracy by focusing computational effort on the most relevant regions
Solution Approach 2:
The patent applies full-search motion estimation only to regions with significant motion or complex patterns, while using simpler estimation methods for static or smoothly varying regions. This localized approach to motion estimation maintains accuracy where needed while reducing overall computational complexity
Data Source
AI summary
Methods and apparatuses are provided for video coding. The method includes: partitioning a picture into a plurality of coding units; partitioning a coding unit of the plurality of coding units into two geometric partitions; and performing a predetermined number of motion vector comparison operations during a process of constructing a merging candidate list for either of the two geometric partitions.