Triangle Prediction Merge Lists for Efficient Video Coding
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 compression methods.
Innovation Solution
Implementing triangular prediction units as a special case of geometric partition prediction in video coding, which involves partitioning video pictures into coding units that are further divided into geometric-shaped prediction units, constructing a merge list of motion vectors, and deriving a uni-prediction merge list for these units under triangular partition mode.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional block-based prediction methods are used, then encoding simplicity is maintained, but coding efficiency deteriorates for high-definition and ultra-high-definition video data
Solution Approach 1:
The patent applies segmentation by dividing a coding unit into multiple prediction units with different geometric shapes (triangular, rectangular, etc.). Each prediction unit can be independently predicted using appropriate reference samples, allowing the encoder to adapt to different motion patterns within the same coding unit. This segmentation improves coding efficiency by reducing prediction errors while maintaining manageable complexity through standardized partitioning rules.
2Loss of information
If geometric partition prediction with triangular prediction units is implemented, then compression performance is improved, but device complexity increases
Solution Approach 1:
The patent implements preliminary action by pre-construction of merge candidate lists containing multiple motion vector candidates from spatial and temporal neighbors before the actual prediction process. This allows the decoder to efficiently select the best prediction mode without performing complex real-time calculations, thereby improving prediction accuracy while controlling device complexity through advance preparation of candidate data.
3Measurement precision
If multiple merge candidates are constructed for triangular prediction units, then prediction accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent applies partial action by constructing a limited set of merge candidates (typically 3-5 candidates) from spatial and temporal neighbors, which is sufficient to achieve good prediction accuracy without exhaustively searching all possible motion vectors. This partial approach balances motion vector accuracy with computational power requirements, avoiding the need for complete motion estimation while still achieving satisfactory compression results.
Data Source
AI summary
Methods and apparatuses are provided for video coding. The method includes: partitioning video pictures into a plurality of coding units (CUs), at least one of which is further partitioned into two prediction units (PUs) including at least one geometric shaped PU; obtaining a first merge list including a plurality of candidates, each including one or more motion vectors; and obtaining a uni-prediction merge list for the geometric shaped PU; where the uni-prediction merge list including a plurality of uni-prediction merge candidates, and each uni-prediction merge candidate includes one motion vector of a corresponding candidate in the first merge list.