Triangular Prediction Units for Video Coding Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies face challenges in efficiently encoding and decoding high-resolution video data while maintaining image quality, particularly with the exponential growth of video data in 4K and 8K resolutions.
Innovation Solution
The implementation of motion compensated prediction using geometric shaped prediction units, such as triangular prediction units, which involve partitioning video pictures into coding units further divided into geometric shaped prediction units, constructing merge lists for motion vectors, and selecting uni-prediction motion vectors based on binary reference list indication flags.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional block-based prediction methods are used, then the encoding process is simple, but coding efficiency deteriorates in high-resolution video scenarios
Solution Approach 1:
The prediction unit is segmented into two triangular regions instead of using traditional rectangular blocks. This segmentation allows the prediction to adapt to directional motion patterns more effectively, improving coding efficiency for high-resolution video by capturing motion gradients within the block
Solution Approach 2:
Different motion characteristics are applied to different regions of the prediction unit. By dividing the block into two triangles with potentially different motion vectors, the method captures local motion variations within the prediction unit, improving prediction accuracy without requiring a complete redesign of the encoding framework
2Adaptability or versatility
If geometric shaped prediction units are implemented, then prediction flexibility improves, but implementation complexity increases
Solution Approach 1:
The prediction unit uses asymmetric triangular shapes instead of symmetric rectangular blocks. This asymmetry allows better adaptation to directional motion patterns and motion gradients in video content, providing improved prediction flexibility while maintaining a relatively simple implementation through basic geometric division
Solution Approach 2:
The geometric shape of the prediction unit is made dynamic by allowing different triangular configurations and motion vector assignments for each partition. This enables the prediction method to adapt to varying motion patterns in different video regions and time instances, enhancing versatility without requiring complex real-time shape optimization
3Measurement precision
If motion vector selection from multiple reference lists is performed, then prediction accuracy improves, but computational complexity increases
Solution Approach 1:
Instead of exhaustively searching all possible motion vectors from multiple reference lists, the method selectively evaluates a subset of candidates. By using triangular partitioning to guide the search and limit the candidate set, the system achieves sufficient motion vector precision while reducing computational power requirements through partial evaluation rather than complete enumeration
Data Source
AI summary
Methods and apparatuses are provided for video coding. The method includes: partitioning video pictures into a plurality of CUs, at least one of which is further partitioned into two PUs including at least one geometric shaped PU; constructing a first merge list including a plurality of candidates, each candidate being a MV including a List 0 MV, and/or a List 1 MV; locating a first candidate for the first PU and a second candidate for the second PU according to a first index and a second index, respectively; obtaining a first uni-prediction MV for the first PU by selecting a List X1 MV of the first candidate according to a first binary reference list indication flag; and obtaining a second uni-prediction MV for the second PU by selecting a List X2 MV of the second candidate according to a second binary reference list indication flag.