Hierarchical Motion Estimation for Video Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing technologies face challenges in accurately estimating motion vectors, especially when dealing with repetitive patterns, leading to misjudgments and degraded viewing experiences. Additionally, there is a lack of effective mechanisms to evaluate the reliability of motion vectors, resulting in increased computational burden and potential image artifacts.
Innovation Solution
The proposed solution involves an image processing apparatus and method that downsizes current and reference frames, performs hierarchical motion estimations to generate motion vectors and reliability values, and uses these reliability values to adjust the motion compensation process, thereby improving accuracy and reducing computational costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If motion estimation is performed on full-resolution frames to detect repetitive patterns, then measurement precision of motion vectors is improved, but computational burden increases significantly
Solution Approach 1:
The patent divides the motion estimation process into multiple stages: first performing motion estimation on down-sized frames to obtain initial motion vectors, then using these as references for refined estimation on full-resolution frames. This segmentation allows the system to balance computational cost and accuracy by processing frames at different resolutions in sequence rather than all at full resolution.
Solution Approach 2:
The patent performs preliminary motion estimation on down-sized frames before performing the final motion estimation on full-resolution frames. This preliminary action provides initial motion vector estimates that guide the subsequent full-resolution estimation, reducing the search space and computational requirements while maintaining accuracy.
2Productivity
If motion compensation is performed without reliability evaluation, then processing speed is maintained, but image quality deteriorates due to misjudgments from repetitive patterns
Solution Approach 1:
The patent introduces a feedback mechanism where motion vectors from down-sized frame estimation are used as reference values for the full-resolution estimation process. This feedback loop allows the system to evaluate and refine motion vector reliability by comparing initial estimates with refined results, reducing misjudgments while maintaining efficient processing.
3Reliability
If full motion estimation is performed on all frames, then reliability of motion vectors is improved, but processing time increases
Solution Approach 1:
The patent applies partial action by performing simplified motion estimation on down-sized frames first, then applying only the necessary refined estimation on full-resolution frames where needed. This approach achieves sufficient reliability without the excessive computational cost of performing complete full-resolution estimation on all frames.
Data Source
AI summary
An image processing apparatus is configured to execute the following steps. A first motion estimation is performed on a down-sized current frame and a down-sized reference frame to generate first motion vectors and a first reliability corresponding to the first motion vectors. An n-th motion estimation is performed on a current frame and a reference frame based on the first motion vectors and the first reliability to generate n-th motion vectors and a n-th reliability corresponding to the n-th motion vectors. A compensated frame between the current frame and the reference frame is generated based on the n-th motion vectors and the n-th reliability.