Geometric Partition Intra Prediction for Video Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies face inefficiencies in intra prediction due to the use of a limited number of prediction directions, which results in suboptimal compression ratios and increased bit usage for less likely directions.
Innovation Solution
Implementing a geometric partition mode (GPM) that divides coding units into rectangular partitions with distinct intra prediction modes, allowing for more flexible and adaptive intra prediction based on neighboring sample directions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional intra prediction with limited prediction directions is used, then device complexity is reduced, but compression ratio deteriorates and bit usage increases for less likely directions
Solution Approach 1:
The patent segments the prediction space by dividing the coding unit into multiple partitions (first partition and second partition) with different prediction directions. Each partition uses a dedicated prediction mode optimized for its specific direction, allowing the system to represent less likely directions with fewer bits by associating them with specific partitions rather than using a single complex prediction mode for all directions.
Solution Approach 2:
The patent applies local quality by assigning different prediction modes to different partitions of the coding unit. Instead of using a uniform prediction mode for the entire block, each partition receives a prediction mode tailored to its local characteristics and direction, improving compression efficiency for each region while the overall system remains manageable through structured partitioning.
2Productivity
If more prediction directions are added to improve compression, then compression ratio improves, but bit usage for less likely directions increases
Solution Approach 1:
The patent segments the prediction space by dividing the coding unit into multiple partitions (first partition and second partition) with different prediction directions. Each partition uses a dedicated prediction mode optimized for its specific direction, allowing the system to represent less likely directions with fewer bits by associating them with specific partitions rather than using a single complex prediction mode for all directions.
Solution Approach 2:
The patent changes the parameter of prediction mode selection based on partition characteristics. By determining which partition (first or second) should use which prediction mode based on factors like partition size, position, or content characteristics, the system optimizes bit usage by selecting the most efficient prediction mode for each specific context rather than using a fixed set of directions for all cases.
Data Source
AI summary
In the method, coded information of a coding unit (CU) in a picture of a video is received from a coded video bitstream. The CU is partitioned into a first partition and a second partition based on a geometric partition mode (GPM). The first partition and the second partition are rectangular partitions that are adjacent to each other and divided by a straight partition line. A first intra prediction mode for the first partition of the CU and a second intra prediction mode for the second partition of the CU are determined. The first intra prediction mode is different from the second intra prediction mode. The first partition of the CU is reconstructed based on the first intra prediction mode and the second partition of the CU is reconstructed based on the second intra prediction mode.