Motion Estimation Algorithm with Adaptive Search and Early Exit
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing motion estimation algorithms in predictive coding schemes are resource-intensive and sacrifice visual quality, failing to consider the bit requirements of motion vectors and often require full search, which is computationally costly.
Innovation Solution
A method for motion estimation that selects initial search points based on error signal energy and cost function using SAD, performs a local search with progressively decreasing step sizes, and allows early exit when a new search point with a lower cost function is found, enabling fast full-pixel and sub-pixel estimation with reduced complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If full search (FS) is performed for block matching, then motion estimation accuracy is improved, but computational resources and time consumption increase significantly
Solution Approach 1:
The search window is divided into multiple regions, and the search process is segmented into multiple passes with different search ranges. In each pass, only specific regions are searched based on the current pass results, reducing the total search space while maintaining accuracy.
Solution Approach 2:
The algorithm performs preliminary actions by establishing initial motion vectors from reference frames and using them to guide subsequent searches. The search window is initialized with predicted motion vectors before the actual search begins, allowing the search to start from informed positions rather than exhaustive scanning.
2Productivity
If algorithms are proposed to reduce full search computational cost, then processing speed is improved, but visual quality is sacrificed
Solution Approach 1:
Different search strategies are applied to different regions of the search window based on local characteristics. Regions with higher motion likelihood are searched more extensively while regions with lower motion likelihood are searched less, optimizing the balance between processing speed and visual quality.
Solution Approach 2:
The search algorithm dynamically adjusts the search range and strategy based on the current frame characteristics, motion vector magnitude, and previous search results. The search window expansion and contraction is adaptive, allowing the algorithm to allocate computational resources dynamically rather than using a fixed search pattern.
3Device complexity
If existing motion estimation algorithms are used, then processing is simplified, but bit requirements for motion vectors are not considered
Solution Approach 1:
The algorithm changes the parameter of motion vector precision dynamically. When the search indicates that higher precision is not necessary, the algorithm uses coarser motion vector representations, reducing the bit requirements for encoding while maintaining acceptable visual quality.
Data Source
AI summary
A motion estimation algorithm finds the best match for a given block or macroblock so that the resulting error signal has very low energy level which is computed, for e.g., by the SAD method. The motion estimation algorithm also provides for an optional sub-pixel level estimation and an inter4v search, and allows for restricting the number of searches for a Frame-frame ME (motion estimation) using Top-Top and Bottom-Bottom field MEs. The algorithm provides for a selective early exit and enables selecting a suitable search area with N candidate points (4 to 8) for starting the search. The search is conducted progressively till a minimum error signal (low energy level signal) is reached. The candidate points for search may be in a diamond shaped configuration, and there may be a plurality of successive diamond configurations, the number of which is configurable. The invention has application in MPEG-4 and H.264 standards.