Timestamp-independent Motion Vector Prediction for Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding techniques, such as the H.264/AVC standard, face inefficiencies in encoding and decoding due to reliance on temporal dependence of motion vector values, which deteriorate with increasing temporal distance between frames, especially in multiple frame referencing scenarios, and lack timestamp independence, affecting coding efficiency and robustness.
Innovation Solution
The method involves establishing reference pictures for current video frames, using motion vector predictors (MVPs) that combine spatial and temporal predictions, allowing for timestamp-independent motion vector calculation, and employing adaptive techniques like Rate Distortion Optimization (RDO) to select between Direct and Skip modes, reducing computational complexity and memory requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If temporal dependence of motion vector values is used for prediction, then coding efficiency is improved for close frames, but performance deteriorates as temporal distance between frames increases
Solution Approach 1:
The patent combines spatial prediction and temporal prediction methods into a unified motion vector prediction system. The motion vector predictor is derived by merging information from spatially adjacent blocks and temporally correlated reference frames, allowing the system to maintain accuracy across varying temporal distances while improving overall coding efficiency.
Solution Approach 2:
The patent introduces adaptive selection between different prediction modes (Direct mode and Skip mode) based on picture type and temporal characteristics. The system dynamically adjusts the prediction strategy depending on whether the current picture is a P or B picture, and selects the most appropriate motion vector predictor from multiple candidates, making the prediction system flexible and adaptive to different temporal scenarios.
2Productivity
If timestamp information is required for motion vector prediction, then temporal statistical dependence can be exploited, but timestamp independence is lost affecting robustness
Solution Approach 1:
The patent extracts and eliminates the dependency on timestamp information from the motion vector prediction process. By deriving motion vector predictors solely from spatial and temporal data available in the reference frames themselves, the system removes the need for external timestamp information, achieving both coding efficiency and timestamp independence for improved robustness.
Solution Approach 2:
The prediction system becomes self-sufficient by using only the information inherently present in the video data (spatial adjacent blocks and temporal reference frames) without requiring external timestamp information. The system serves itself by deriving all necessary prediction parameters from the video content alone, enhancing robustness and adaptability.
3Adaptability or versatility
If multiple picture referencing is enabled, then coding flexibility is improved, but problems with temporal dependence and timestamp requirements increase
Solution Approach 1:
The patent segments the motion vector prediction process into distinct modes (Direct mode for P pictures, Skip mode for B pictures) with specific selection criteria for each. This segmentation simplifies the overall complexity by providing clear, rule-based decision paths for different picture types, making the multiple picture referencing system more manageable while maintaining flexibility.
Solution Approach 2:
The patent changes the prediction parameters and selection logic based on picture type (P or B pictures). By adapting the motion vector predictor selection and prediction mode according to the specific picture type and temporal characteristics, the system maintains flexibility for multiple picture referencing while managing complexity through parameter-based adaptation rather than complex universal logic.
Data Source
AI summary
Methods and apparatuses are provided for achieving improved video coding efficiency through the use of Motion Vector Predictors (MVPs) for the encoding or decoding of motion parameters within the calculation of the motion information in B pictures and/or P pictures. Certain exemplary methods and apparatuses selectively apply temporal and/or spatial prediction. Rate Distortion Optimization (RDO) techniques are also applied in certain methods and apparatuses to further help improve coding efficiency.