Video Compression Tagging to Cut External Memory Bus Load
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing video compression process in baseboard management control chips requires multiple reads and writes to an external storage apparatus, leading to high bus occupancy and reduced performance for other applications.
Innovation Solution
Implement frame-wise writing of original video data to an external storage apparatus and line-by-line comparison to identify identical data, converting identical data into tag information and storing it in internal storage, while non-identical data is directly stored in external storage, reducing redundant reads and writes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If video compression processing is performed by reading and writing original video data multiple times to external storage apparatus, then video compression function is achieved, but bus occupancy rate increases and system performance decreases
Solution Approach 1:
The patent divides video data processing into two segments: identical data (handled via tags and bookmarks) and non-identical data (handled via traditional read-write). This segmentation allows the system to process only the necessary portions of video data, reducing bus occupancy and improving system performance while maintaining video compression functionality.
Solution Approach 2:
The patent performs preliminary comparison between current frame and previous frame to identify identical data before compression processing. By pre-identifying identical data and replacing it with tags, the system avoids redundant read-write operations during the main compression process, thereby reducing bus occupancy and enhancing overall system performance.
2Reliability
If all original video data is written to external storage apparatus for compression processing, then complete video data is available for processing, but read-write load on external storage increases
Solution Approach 1:
The patent extracts identical data from the video processing stream and replaces it with compact tag information. By taking out the redundant identical data and substituting it with minimal tag representations, the system maintains video data completeness for compression processing while significantly reducing the read-write load on external storage apparatus.
Solution Approach 2:
The patent changes the representation parameter of identical data from full video data to compact tag information. This parameter change allows the system to maintain the ability to reconstruct complete video data while dramatically reducing the amount of data that needs to be read and written to external storage, thereby reducing read-write load.
Data Source
AI summary
Disclosed are a method and apparatus for processing video compression. The method includes: controlling a comparison apparatus to perform frame-wise writing of original video data to an external storage apparatus, and controlling a video compression control component to perform frame-wise reading of the original video data from the external storage apparatus; in response to writing original video data of a non-first frame into the external storage apparatus, and in response to writing the original video data of the current frame into the external storage apparatus, converting the video data into tag information and writing the tag information into the external storage apparatus; and in response to reading the original video data of the previous frame from the external storage apparatus, writing the video data corresponding to the current same line number into an internal storage apparatus.