NVIDIA Metropolis
An application framework that simplifies the development, deployment and scale of AI-enabled video analytics applications from edge to cloud.
NVIDIA Metropolis is an end-to-end application framework that includes pre-trained models, training and optimization tools, deployment SDKs, CUDA-X libraries and the NVIDIA EGX platform.
The NVIDIA Metropolis application framework makes it easier for developers to combine common video cameras and sensors with AI-enabled video analytics to provide operational efficiency and safety applications across a broad range of industries - including retail analytics, city traffic management, airport operations and automated factory inspections.
Applications are optimized to run on a wide range NVIDIA EGX™ platform hardware, allowing developers to deploy anywhere securely, from the edge to the cloud.
Faster Time to Solution
Leverage high-performance, pre-trained models to build accurate AI applications across a broad range of industries.
Greater Throughput
Experience production ready models and SDKs to reduce development costs and achieve high throughput inference.
Secure, Scalable Deployment
Manage AI deployments securely and scalably from dozens up to millions of servers or edge devices.
Powerful AI Tools for
Intelligent Video Analytics
Pre-Trained Models
Eliminate the time-consuming process of building models from scratch. Choose from over 100+ permutations of highly accurate models and generic neural network architectures.
Transfer Learning Toolkit
AI training toolkit that abstracts away the AI/DL framework complexity. Build production quality pre-trained models faster with no coding required.
DeepStream SDK
The streaming analytics toolkit for AI-based multi-sensor processing, video, audio, and image understanding. Build seamless high performance apps and services from edge to cloud.
Triton Inference Server
The open source inference serving software that simplifies the deployment of AI models at scale in production.
TensorRT
An SDK for high-performance deep learning inference which includes an inference optimizer and runtime that delivers low latency and high throughput at edge/embedded.
Video Codec SDK
Comprehensive set of APIs, including high-performance tools, samples and documentation for hardware-accelerated video encode and decode on Windows and Linux.
Achieve Deep Insights With Seamless AI Video Analytics Applications & Services
Get StartedProduction Ready Models and SDKs for Computer Vision
Speed up your prototype to production journey with NVIDIA Metropolis. Use our readily available, production-grade pre-trained models for vehicle detection, gesture recognition, object classification, and more. Utilize SDKs that provide a complete video streaming analytics toolkit for AI-based multi-sensor processing.
End-to-End AI Solution With Over 10X Development Speedup
Build end-to-end services and solutions for transforming pixels and sensor data to actionable insights. You can easily integrate the production-ready AI models produced by TLT with NVIDIA DeepStream SDK and TensorRT to unlock greater performance and high throughput inference.
Developer blogs: Implementing a Real-time Face Mask Detector Application for COVID-19, Creating a Real-Time License Plate Detection and Recognition App
Secure Application Management With Fleet Command
Seamlessly manage, scale, and keep your entire AI deployment running smoothly with NVIDIA Fleet Command. Fleet Command is a hybrid-cloud platform for securely and remotely deploying, managing, and scaling AI applications across dozens or up to millions of servers or edge devices. The streamlined provisioning, detailed monitoring dashboard, and extensive automated processes allow your IT and system administrators team of any skill level to manage the deployment and maintenance of AI applications at the edge.
Leading Adopters Across All Verticals
Resources
Train Smarter with Pre-Trained Models and TLT
Learn how the world’s top AI teams combine pre-trained models and transfer learning tools to supercharge their AI vision development.
WATCH NOW
Preview of DeepStream SDK 6.0
Watch the preview of the next version of DeepStream SDK and explore the new Graphical User Interface (GUI) and rich set of productivity capabilities.
WATCH NOW
How Cities are Turning AI into Cost Savings
Learn how the City of Raleigh,North Carolina partners with ESRI to use real-time roadway insights from AI-powered video analytics to reduce operation costs.
WATCH NOW
How AI Saves Lives and Money for Railways
Discover how AI-based video analytics solutions can be used to save money and increase operational efficiency in Metro rail networks.
WATCH NOW
Developer Blog
Find out the latest about data generation and annotation tools for NVIDIA Transfer Learning Toolkit.
READ BLOG
Developer Tutorial
Learn how to train state-of-the-art models for classification and object detection.
READ BLOG
Developer Webinar
Learn how to create a gesture recognition application with robot interactions.
WATCH NOW ACCESS TO GITHUB REPO
Developer Tutorial
Learn how to create a real-time number plate detection and recognition app.
WATCH NOW