Temporal Motion Vector Prediction for Variable Picture Size Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies face challenges in efficiently processing changes in picture size, leading to inefficiencies in temporal motion vector prediction due to the unavailability of reference pictures with the same size, which affects encoding and decoding performance.
Innovation Solution
An encoder and decoder system that enables or disables temporal motion vector prediction based on the availability of reference pictures with the same size, allowing for efficient encoding and decoding by encoding/decoding a parameter indicating whether a change in picture size is allowed, and disabling temporal motion vector prediction when necessary.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If temporal motion vector prediction is used when reference pictures with the same size are unavailable, then motion vector prediction can be performed, but coding efficiency deteriorates due to inaccurate prediction
Solution Approach 1:
The patent dynamically adjusts the temporal motion vector prediction functionality based on picture size availability. When reference pictures with the same size are unavailable, the system disables temporal motion vector prediction to prevent inaccurate predictions. This dynamic control mechanism ensures that prediction operations are only performed when conditions are favorable, thereby maintaining both prediction accuracy and coding efficiency.
2Adaptability or versatility
If picture size changes are allowed in video sequences, then adaptability improves, but processing complexity increases due to reference picture availability issues
Solution Approach 1:
The patent applies local quality by treating different picture size scenarios differently. When picture sizes change, the system locally disables temporal motion vector prediction for affected picture pairs while maintaining it for others where size matching is available. This selective approach allows picture size adaptability without uniformly increasing processing complexity across all scenarios.
Solution Approach 2:
The system performs preliminary checks to determine whether reference pictures with the same size are available before attempting temporal motion vector prediction. By pre-assessing picture size compatibility, the system avoids unnecessary processing complexity that would arise from attempting predictions on incompatible picture pairs, thus enabling picture size changes while managing processing load.
3Productivity
If temporal motion vector prediction is disabled when reference pictures are unavailable, then coding efficiency improves by avoiding redundant processing, but prediction functionality is reduced
Solution Approach 1:
The patent implements dynamic control of temporal motion vector prediction based on real-time assessment of reference picture availability. The system enables the prediction functionality when suitable reference pictures exist and disables it only when picture size mismatches occur. This dynamic approach maximizes coding efficiency by avoiding redundant processing while preserving prediction functionality wherever possible.
Data Source
AI summary
An encoder includes circuitry and memory coupled to the circuitry. In operation, the circuitry encodes, into a sequence parameter set, a first parameter indicating that a change in a picture size is allowed for any of pictures, determines whether or not a reference picture having a same size as a current picture is available to encode the current picture, and disables temporal motion vector prediction when it is determined that the reference picture having the same size as the current picture is not available.