Edge Computing Unit for Video Subject Query
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing cloud computing architectures face challenges in latency, availability, bandwidth usage, data privacy, network security, and the capacity to process large volumes of data in real-time, especially in remote operating environments with limited infrastructure.
Innovation Solution
The implementation of edge computing units that can process and convert digital video data into natural language text descriptions, allowing for real-time or near-real-time processing and storage of data at edge locations, and providing question and answer sessions based on video content.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If sensor data is transmitted over satellite communication links from remote operating environments, then data can be transmitted without infrastructure requirements, but transmission speed is extremely slow and cost is high
Solution Approach 1:
The patent segments the data processing function into two parts: local processing at the edge computing unit for immediate needs, and cloud processing for comprehensive analysis. This segmentation allows critical data to be processed locally without satellite transmission delays, while only essential processed data or alerts are transmitted via satellite, resolving the contradiction between remote accessibility and transmission speed.
Solution Approach 2:
The edge computing unit acts as an intermediary between remote sensors and satellite communication systems. It pre-processes sensor data locally, filtering and consolidating information before transmission, thereby reducing the volume of data that must traverse the slow satellite link while maintaining the ability to operate in remote environments without infrastructure.
2Reliability
If all raw sensor data is stored at the edge location, then complete data is available for analysis, but storage requirements become prohibitively large
Solution Approach 1:
The edge computing unit extracts only the essential features and processed results from raw sensor data, storing these extracted insights locally while transmitting comprehensive data to the cloud for long-term archival. This extraction approach maintains data reliability for local operations without requiring prohibitively large local storage capacity.
Solution Approach 2:
The patent transitions data from one dimension (local storage) to another (cloud storage) based on data type and usage requirements. Critical processed data remains locally stored, while raw and historical data are migrated to cloud storage infrastructure, effectively managing storage requirements across different dimensional spaces.
3Speed
If video data is transmitted over conventional fiber optic links, then transmission speed is high, but infrastructure availability is limited in remote areas
Solution Approach 1:
The edge computing unit provides self-service by performing local data processing and analysis, reducing dependency on external communication infrastructure. It can autonomously process sensor data and generate alerts without requiring high-speed fiber optic connections, thereby maintaining operational speed and flexibility in remote locations where fiber infrastructure is unavailable.
4Productivity
If centralized cloud processing is used, then processing capacity is high, but latency increases and data privacy concerns arise
Solution Approach 1:
The edge computing unit performs preliminary data processing, filtering, and analysis before data is transmitted to or processed by centralized cloud systems. This preliminary action reduces the volume of data requiring centralized processing and enables faster local responses to time-critical events, thereby maintaining high productivity while minimizing latency for urgent operations.
Data Source
AI summary
Disclosed are systems and methods that convert digital video data, such as two-dimensional digital video data, into a natural language text description describing the subject matter represented in the video. For example, the disclosed implementations may process video data in real-time, near real-time, or after the video data is created and generate a text-based video narrative describing the subject matter of the video. In addition, the disclosed implementations may also support a question and answer session in which a user may submit queries about the subject matter of one or more videos and the disclosed implementations will present natural language responses based on the subject matter of the video and any corresponding context.