Computational Storage Using Inference Video and Event Tables
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The inefficiency in managing storage space for multimedia data in computational storage systems limits the performance of electronic devices, as high-capacity multimedia data requires significant storage space and can lead to performance deterioration.
Innovation Solution
A computational storage system that utilizes neural network models, base data, and event tables to generate inference video data, allowing efficient storage by replacing high-capacity original video data with lower-capacity data sets, thereby optimizing storage space utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If original video data is stored in the storage device, then data quality is maintained, but storage space is excessively consumed
Solution Approach 1:
The patent creates a compressed copy of the original video data through neural network inference. The computing device generates inference video data from base data and event tables, which are derived from the original video. This inferred version retains essential information while occupying significantly less storage space, effectively replacing the need to store the complete original video data.
Solution Approach 2:
The patent extracts only the critical information from the original video data into base data and event tables. By separating and extracting only the necessary features and temporal information, the system reduces storage requirements while maintaining data quality for the purposes of computational storage operations.
2Productivity
If multimedia data is stored without efficient management, then data accessibility is maintained, but system performance deteriorates due to storage space constraints
Solution Approach 1:
The patent performs preliminary processing of video data by generating base data and event tables before storage. The neural network model pre-processes the original video data into a compressed representation that can be efficiently stored and accessed later, preparing the data in advance to avoid performance issues during actual storage operations.
Solution Approach 2:
The patent changes the data representation parameters by transforming original video data into a different format consisting of base data and event tables. This parameter transformation reduces the data size and improves storage efficiency while maintaining accessibility and system performance.
Data Source
AI summary
An example computational storage system includes a storage device and a computing device. The computing device is configured to generate first inference multimedia data corresponding to original multimedia data based on base data, an event table, and at least one neural network model, where the base data includes base raw data of at least one object included in the original multimedia data, and the event table includes data obtained based on respectively mapping events occurred in the base data and occurrence times of the events.