Key Features:
-
RTSP Camera Capture: Connect and manage network camera streams for real-time video analytics.
-
DeepStream AI Integration: Run inference using NVIDIA DeepStream and custom AI models.
-
Event Logging: Record detected events and relevant application information to a dashboard.
-
Event Snapshots: Capture images when predefined events or actions occur.
-
Custom Business Rules: Define policies that filter inference results and determine how the application should respond.
-
Custom Actions: Trigger application-specific actions based on detected events.
-
Media Processing Interfaces: Extend or customize the video processing workflow for different application requirements.
-
Modular Architecture: Replace individual building blocks without redesigning the entire application.
How It Works:
The framework is organized around four main components:
-
Camera Capture: Controls the video sources used by the application.
-
AI Manager: Processes DeepStream inference and forwards detection information.
-
Action Dispatcher: Uses application-defined policies to determine which actions should be executed.
-
Config Parser: Loads the parameters required to configure the application before execution.
Developers can define their own camera source, DeepStream model, policies, actions, and custom media processing while reusing RidgeRun’s underlying framework.
Included Reference Design:
The product includes the Automatic Parking Lot Vehicle Registration reference design.
This example demonstrates how DeepStream inference can be combined with business rules and application actions to identify predefined events and record information such as:
-
Time
-
Date
-
Location
-
Event-related actions
It provides a practical starting point for building more specialized AI video analytics applications.
Example Use Case:
A parking facility needs an AI-based system to analyze camera streams and identify vehicle-related events.
Using the DeepStream Reference Design, the system can capture RTSP streams, run vehicle detection through DeepStream, apply custom business rules, capture event snapshots, and log relevant information to a dashboard.
The same framework can then be adapted for other applications by replacing the AI model, policies, and actions.
Perfect For:
-
Smart parking systems
-
Video surveillance
-
Smart cities
-
Industrial monitoring
-
Retail analytics
-
Traffic monitoring
-
Robotics and autonomous systems
-
Custom edge AI applications
What’s Included:
-
Python-based DeepStream Reference Design
-
Automatic Parking Lot Vehicle Registration example
-
RTSP camera capture
-
Event logging and snapshots
-
Custom media processing interfaces
-
Business rules interface
-
Custom action framework
-
5 hours of engineering support
Supported Platforms:
-
NVIDIA Jetson Orin
-
NVIDIA Jetson Orin Nano
-
NVIDIA Jetson Orin NX
-
NVIDIA Jetson Xavier
-
NVIDIA Jetson Xavier NX
-
NVIDIA Jetson TX2
-
NVIDIA Jetson Nano
-
x86 systems with NVIDIA GPU
Learn more about the architecture, customization, inference parsers, policies, actions, and media processing on our Developer Wiki:
Monthly Subscription
Hire RidgeRun at a flat monthly rate for a maximum of 8 hours per day, 40 hours per week each month. Get fast interaction with engineers working on your project, weekly 45 minute calls to update you on our progress, and the complete engagement and expertise of our engineering team.
Work With UsFAQs
It is a modular, Python-based framework for building video analytics applications with the NVIDIA DeepStream SDK.
It provides the application infrastructure required to capture video, process AI inference results, apply business rules, execute actions, and record events without developing every component from scratch.
Yes. Developers can define actions that run when the configured policies and conditions are satisfied.
Customer Reviews
“I must say thus far yourself and the entire RidgeRun team have been a pleasure to work with. I realize what we have been asking for is on the demanding end of the spectrum. I am genuinely appreciative for all of the assistance you’ve provided us so far and have shown a notable amount of good-will in this effort. It has not gone unnoticed.”
Customer industry:
AudioVisual Devices
Customer Reviews
“It's been an absolute pleasure working with you, and everyone there at RidgeRun continues to impress us here. I've worked with many contractors and companies over my career and you are by far the most proficient and knowledgeable.”
Customer industry:
Drones
Customer Reviews
"As a side note: just to let you know, I'm really happy to be working with RidgeRun in general. It's pretty clear that you tell it like you see it, and know what you do and don't know; it's really refreshing. That's super rare on these kinds of technical engagements when everyone seems to either want to claim expertise they don't have, or are so cautious they are don't actually say anything useful"
Customer industry:
Drones and Robotics
Customer Reviews
"RidgeRun has been so good that part of me doesn't even want to write this testimonial—because if more companies discover them, I might have to share the amazing engineers I get to work with. But credit where credit's due: their team is sharp, responsive, and consistently delivers beyond expectations. They make tough technical challenges feel easy, and that's rare."
Ayman
Director of Software