Dynamic Video Encoder Search Window Reload
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional video compression methods using motion-compensated prediction face inefficiencies due to the limited size of the search window, which fails to capture data that migrates outside the sliding window, leading to less accurate predictions and increased power consumption during window reloading.
Innovation Solution
The method dynamically determines locations to reload the motion estimation search window based on a threshold number of reloads, using a motion map to identify areas where the window should be reloaded for more accurate prediction, balancing efficiency and accuracy by sliding the window in a direction indicated by the next co-located portion for remaining locations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If a smaller search window is used, then local memory size is reduced, but prediction accuracy deteriorates when data migrates outside the window
Solution Approach 1:
The patent implements dynamic search window reloading based on motion map analysis. The search window is selectively reloaded at locations where objects are likely to migrate outside the window boundaries, making the memory usage adaptive rather than static. This resolves the contradiction by maintaining small memory size overall while dynamically expanding coverage only where needed for accuracy.
Solution Approach 2:
The patent performs preliminary analysis using a motion map to predict object migration patterns before conducting the actual search. This preliminary action identifies which locations require search window reloading, allowing the system to proactively prepare memory allocations only where migration is expected, thus balancing memory efficiency with prediction accuracy.
2Measurement precision
If a larger search window is used, then prediction accuracy is improved, but local memory size increases
Solution Approach 1:
The patent applies different memory allocation strategies to different regions of the search space. Rather than uniformly increasing memory for the entire search window, it selectively reloads memory only at specific locations where object migration is predicted based on motion map analysis. This local quality approach maintains small overall memory size while ensuring accuracy where needed.
3Measurement precision
If the search window is reloaded frequently, then prediction accuracy for migrating data is improved, but power consumption increases
Solution Approach 1:
The patent uses motion map analysis as a feedback mechanism to guide search window reloading decisions. By analyzing motion patterns and predicting where objects will migrate, the system intelligently determines which locations require reloading, avoiding unnecessary memory operations and reducing power consumption while maintaining prediction accuracy where objects are likely to move.
4Use of energy by moving object
If the search window is slid without reloading, then power consumption is reduced, but data migration outside the window is not captured
Solution Approach 1:
The patent implements a dynamic hybrid approach that combines sliding window efficiency with selective reloading. The search window slides to maintain efficiency, but at locations where motion map analysis predicts object migration, the window is reloaded instead of merely slid. This dynamic adaptation resolves the contradiction by switching between sliding and reloading based on actual migration needs.
Data Source
AI summary
A processing apparatus is provided that includes an encoder configured to encode current frames of video data using previously encoded reference frames and perform motion searches within a search window about each of a plurality of co-located portions of a reference frame. The processing apparatus also includes a processor configured to determine, prior to performing the motion searches, which locations of the reference frame to reload the search window according to a threshold number of search window reloads using predicted motions of portions of the reference frame corresponding to each of the locations. The processor is also configured to cause the encoder to reload the search window at the determined locations of the reference frame and, for each of the remaining locations of the reference frame, slide the search window in a first direction indicated by the location of the next co-located portion of the reference frame.