GPU Appliance for Video Signature Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional surveillance systems face insufficient GPU processing power for high-performance semantic analytics in video surveillance, leading to inefficiencies as the load increases with advanced video analytics.
Innovation Solution
A surveillance system architecture that includes a dedicated GPU appliance shared among Network Video Recorders (NVRs) and analytics appliances, utilizing multiple GPU cards to distribute and scale processing power, allowing for efficient generation and processing of signatures from video frames across a network.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Power
If traditional surveillance systems are used with standard processing capabilities, then device complexity is kept low, but GPU processing power becomes insufficient for high-performance semantic analytics
Solution Approach 1:
The system divides the surveillance infrastructure into separate functional components: existing NVRs that continue to handle video recording and basic functions, and new GPU appliances that provide dedicated high-performance processing power for semantic analytics. This segmentation allows the system to scale processing power without increasing complexity across the entire system.
Solution Approach 2:
The GPU appliances serve multiple purposes: they process semantic analytics for multiple NVRs simultaneously, generate signatures for various video feeds, and can be shared across different surveillance applications. This multi-functionality maximizes the utilization of added processing power while avoiding proportional increases in overall system complexity.
2Productivity
If advanced video analytics are implemented to increase productivity, then semantic analytics performance improves, but the load on existing systems increases causing inefficiencies
Solution Approach 1:
The patent extracts the computationally intensive semantic analytics processing from the existing NVR infrastructure and relocates it to dedicated GPU appliances. This extraction removes the processing load bottleneck from the original system, allowing advanced analytics to run at high speed without causing delays or inefficiencies in the core surveillance operations.
Solution Approach 2:
The GPU appliances act as intermediary processing nodes between the video sources and the NVRs. They receive video data, perform complex semantic analytics processing, generate signatures, and return results to the NVRs. This intermediary layer handles the heavy computational load separately, preventing it from bottlenecking the overall system and maintaining processing efficiency.
3Productivity
If GPU processing power is increased to meet video analytics demands, then signature generation efficiency improves, but network disruption may occur
Solution Approach 1:
The system implements dynamic load distribution where GPU appliances can adaptively handle varying analytics demands from multiple NVRs. The architecture allows flexible allocation of processing resources based on current network conditions and analytics requirements, enabling efficient signature generation while maintaining network stability through adaptive resource management.
Data Source
AI summary
Generating signatures within a network that includes a plurality of computing devices of varying processing capabilities is disclosed. Chips may be transmitted, from a network video recorder and over the network, to an analytics appliance having a GPU processing power that is higher than a GPU processing power possessed by the network video recorder. The GPU processing power possessed by the analytics appliance may be employed to process the chips therein and generate respective signatures.