Weighted Motion Prediction for Inter-Intra Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video compression technologies, such as HEVC and VVC, face inefficiencies in combining inter and intra prediction modes, leading to inconsistent designs and sub-optimal performance in video encoding and decoding processes.
Innovation Solution
Implementing a method for video encoding and decoding that combines multiple motion prediction modes with corresponding weights for sub-blocks, using triangular and multi-shape predictions, and adapting the number of regions based on block size, while restricting certain prediction modes to unify the design and improve efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple motion prediction modes are combined with corresponding weights for sub-blocks, then video compression efficiency is improved, but device complexity increases
Solution Approach 1:
The current block is divided into multiple sub-blocks, and different prediction modes are applied to different sub-blocks. This segmentation allows the system to achieve better compression efficiency by adapting to local variations in the video content while maintaining manageable complexity through localized processing.
Solution Approach 2:
Different prediction modes and weights are applied to different sub-blocks based on local characteristics of the video content. This local quality approach optimizes compression efficiency for each region while avoiding the need to process the entire block with the most complex mode, thus balancing performance and complexity.
2Measurement precision
If triangular and multi-shape predictions are implemented, then prediction accuracy is improved, but manufacturing precision requirements increase
Solution Approach 1:
The video block is segmented into triangular and multi-shape regions to improve prediction accuracy. These geometric partitions allow the prediction to better follow the actual motion patterns and boundaries in the video content, achieving higher precision without requiring excessive computational resources.
Solution Approach 2:
The prediction structure adapts dynamically by selecting different shapes (triangular, rectangular, etc.) and numbers of regions based on the block size and content characteristics. This dynamic adaptation allows the system to achieve high prediction accuracy while avoiding the need for fixed, overly precise partitioning rules.
3Adaptability or versatility
If the number of regions is adapted based on block size, then adaptability is improved, but device complexity increases
Solution Approach 1:
The number of prediction regions is dynamically adjusted based on the block size and content characteristics. For larger blocks, more regions may be used to capture local variations, while smaller blocks use fewer regions. This dynamic approach improves adaptability without requiring a fixed, complex region management system.
Solution Approach 2:
The system changes key parameters such as the number of regions and their shapes based on the block size and prediction requirements. This parameter adaptation allows the system to be versatile across different video content types and resolutions while maintaining manageable complexity through systematic parameter selection rules.
4Stability of the object's composition
If certain prediction modes are restricted, then design consistency is improved, but productivity decreases
Solution Approach 1:
Certain prediction modes are extracted or removed from the available options to simplify the design and improve consistency. By eliminating modes that cause design inconsistencies or implementation difficulties, the system achieves a more unified and maintainable architecture while retaining the essential prediction capabilities needed for high efficiency.
Data Source
AI summary
At least a method and an apparatus are provided for efficiently encoding or decoding video. For example, a plurality of different motion prediction modes for a current block are obtained. The current block is encoded or decoded based on a combination of the plurality of different motion prediction modes with corresponding weights for a plurality of sub-blocks of the current block, wherein the combination with the corresponding weights comprising an inter prediction mode and an intra prediction mode.