Motion Vector Prediction Using Scaling and Differential Determination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional video encoding methods fail to effectively consider the correlation and similarity among motion vectors when predicting motion vectors for current blocks, leading to inefficient compression and encoding of high-quality digital video content.
Innovation Solution
An image encoding/decoding apparatus and method that predicts motion vectors using vector information from adjacent blocks and previous motion vectors, incorporating scaling and differential determination units to select and scale reference information based on similarity, and employs multiple prediction schemes to accurately estimate motion vectors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional median prediction method is used for motion vectors, then encoding simplicity is maintained, but motion vector prediction accuracy deteriorates due to failure to consider correlation and similarity among motion vectors
Solution Approach 1:
The patent changes the prediction parameter from simple median calculation to a more sophisticated method that considers correlation and similarity among motion vectors. It introduces a determination unit that selects between different prediction schemes (first scheme using median of adjacent block motion vectors, second scheme using correlation-based prediction) based on similarity metrics, thereby improving prediction accuracy while managing complexity through adaptive parameter selection.
2Measurement precision
If multiple motion vectors from adjacent blocks are considered, then prediction accuracy improves, but computational complexity increases
Solution Approach 1:
The patent implements a dynamic prediction system that adapts between different prediction schemes based on real-time similarity assessment. The determination unit dynamically selects whether to use the first prediction scheme (simple median) or second prediction scheme (correlation-based) by evaluating similarity metrics between current and reference blocks, allowing the system to optimize computational power consumption by using simpler methods when appropriate and more accurate methods when needed.
3Measurement precision
If scaling of motion vectors is performed, then prediction accuracy for different reference pictures improves, but processing complexity increases
Solution Approach 1:
The patent applies preliminary scaling to motion vectors based on temporal distance between reference pictures. The scaling unit pre-processes motion vectors by applying scaling factors derived from picture order counts and reference picture indices before they are used in prediction. This preliminary action ensures that motion vectors are appropriately adjusted for different temporal distances, improving prediction accuracy while managing complexity through pre-calculated scaling factors.
Data Source
AI summary
Provided is an image encoding/decoding apparatus and method. The image encoding apparatus may include a motion vector prediction unit to perform a prediction with respect to an arbitrary motion vector of a current block within an image, using at least one of vector information of a motion vector corresponding to an adjacent block and vector information of a previous motion vector of the current block, and a differential determination unit to determine differential information of a motion vector of the current block based on the motion vector predicted in the motion vector prediction unit and an actual motion vector of the current block.