Fast Motion Estimation for Video Encoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video compression standards like H.264|MPEG-4 AVC require computationally intensive motion estimation (ME) processes, which are time-consuming and not fast enough for real-time implementations, especially due to the complexity of exploring multiple motion vectors and partitions.
Innovation Solution
A fast motion estimation process that prioritizes the base reference frame for initial optimization, reduces computational demand by simplifying searches on non-base reference frames, and selectively transmits the final partition and associated motion vector, thereby enhancing processing speed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If full search motion estimation is performed to find the best motion vector for each partition, then motion estimation accuracy is improved, but computational complexity increases significantly
Solution Approach 1:
The patent segments the reference frames into a base layer and an enhancement layer, performing motion estimation in two stages. First, motion vectors are estimated for the base layer using a simplified search, then refinement is performed only for the enhancement layer. This segmentation reduces the overall computational complexity while maintaining accuracy by dividing the search space into manageable parts.
Solution Approach 2:
The patent performs preliminary motion estimation on the base layer before refining with the enhancement layer. By conducting an initial search to obtain approximate motion vectors and then performing refinement only on the enhancement layer, the system avoids performing full searches on all layers, thereby reducing computational complexity while preserving estimation accuracy.
2Adaptability or versatility
If multiple reference frames with multiple partitions are supported to improve video compression flexibility, then adaptability is improved, but motion estimation time increases
Solution Approach 1:
The patent divides reference frames into base and enhancement layers, allowing the system to handle multiple reference frames and partitions efficiently. By segmenting the processing into two stages (base layer estimation followed by enhancement layer refinement), the system maintains flexibility in handling various partition configurations while reducing the time required for motion estimation across multiple references.
Solution Approach 2:
The patent performs full motion estimation only on the base layer, then applies partial refinement only to the enhancement layer. This partial action approach maintains the flexibility of supporting multiple reference frames and partitions while avoiding the excessive computational time that would result from performing full searches on all layers and partitions.
3Measurement precision
If traditional motion estimation processes are used in real-time encoders, then motion compensation accuracy is maintained, but processing speed becomes insufficient
Solution Approach 1:
The patent performs preliminary motion estimation on the base layer to obtain initial motion vectors, then refines these vectors using the enhancement layer. This preliminary action allows the system to achieve accurate motion compensation while significantly reducing processing time compared to traditional methods that perform full searches, thereby meeting real-time encoder requirements.
Solution Approach 2:
The patent extracts the essential motion information from the base layer first, then adds refinement information from the enhancement layer. By separating the motion estimation into two distinct stages and extracting key information in sequence, the system maintains compensation accuracy while achieving the processing speed necessary for real-time operation.
Data Source
AI summary
A process performs a search on a base reference frame in a video signal to determine a first motion vector for each of the possible partitions of a current macroblock. The first motion vector for a partition provides the most optimal match for the partition in the base reference frame. An optimal partition for the current macroblock, which gives a minimum motion estimation cost among all the possible partitions in the base reference frame, is determined. The process performs the search on the optimal partition over a non-base reference frame in the video signal to determine a second motion vector that provides the most optimal match for the corresponding partition in the non-base reference frame. A simplified search is performed on each partition other than the optimal partition over the non-base reference frame to determine the second motion vector for the corresponding partition.