Get Started With NVIDIA Metropolis Microservices
Metropolis microservices give you powerful, customizable, cloud-native building blocks for developing vision AI applications and solutions—built to run on NVIDIA Cloud and data center GPUs.
Release Highlights
This latest release of NVIDIA Metropolis microservices (v2) for enterprise GPUs features:
- New reference workflows and guides on using digital twin and synthetic data generation to improve model robustness and facilitate end-to-end application development, tuning, and validation.
- Advanced new transformer-based detection and ReID models to significantly boost accuracy and robustness in varied tracking scenarios.
- The new Real-Time Location System (RTLS) mode in multi-camera tracking for real-time, accurate location updates.
- Enhanced overall system performance with the introduction of Single-View 3D Tracking (SV3DT) and improved visualization tools across single- and multi-camera tracking environments.
- Upgraded underlying systems to support dynamic configurations, including more efficient Kafka message consumption and the ability to update configurations and calibrations on the fly.
- Edge-to-cloud connectivity features for seamless streaming and inferencing using WebRTC and OpenVPN, alongside dynamic camera stream management in Docker and Kubernetes deployments.
- Enhanced user interfaces with new visualization options, including thumbnails for event cards and polygonal field of view drawings, improving interaction and accessibility.
- Critical bug fixes across various components to enhance stability and user experience, with updates to memory management and UI responsiveness.
Featured Resources
Optimizing End-to-End Vision Workflow Development
The multi-camera tracking reference workflow brings the entire development pipeline from data generation to model training to application development. It helps developers build complex vision AI applications for large spaces.
Read the BlogEnhance Multi-Camera Tracking Accuracy by Fine-Tuning AI Models with Synthetic Data
Learn how to fine-tune a re-ID model with synthetic data to improve the accuracy and reliability of your multi-camera tracking system.
Read the BlogReal-Time Vision AI From Digital Twins to Cloud-Native Deployment
Learn how to develop and deploy vision AI applications with Metropolis microservices and workflows.
Read the BlogGet Started Video Tutorial
Additional Resources
Blogs
- Optimizing End-to-End Vision Workflow Development
- Enhance Multi-Camera Tracking Accuracy by Fine-Tuning AI Models with Synthetic Data
- Real-Time Vision AI From Digital Twins to Cloud-Native Deployment
- Simplifying Camera Calibration to Enhance AI-Powered Multi-Camera Tracking
- NVIDIA Combines Digital Twins With Real-Time AI for Industrial Automation
Documentation and Forum
- Download the Software
- Quick Start Guide
- Docuemntation
- End-to-end Workflow Guide
- Forum ( Please note: Forum access for technical questions requires prior application approval to access the software )
Ethical AI |
NVIDIA platforms and application frameworks enable developers to build a wide array of AI applications. Consider potential algorithmic bias when choosing or creating the models being deployed. Also, work with the model’s developer to ensure that it meets the requirements for the relevant industry and use case; that the necessary instruction and documentation are provided to understand error rates, confidence intervals, and results; and that the model is being used under the conditions and in the manner intended. |