NVIDIA has partnered with Microsoft Azure IoT in transforming and enabling advanced AI innovations for our developers and customers, by making DeepStream; the multi-purpose streaming analytics SDK available on Azure IoT Edge Marketplace.
DeepStream enables a broad set of use cases and industries, to unlock the power of NVIDIA GPUs for smart retail and warehouse operations management, parking management, optical inspection and traffic planning and more. The SDK is a core part of NVIDIA Metropolis; the application framework for building and deploying intelligent video analytics based applications.
Whether you are considering a tiny NVIDIA Jetson Nano, which can deliver one-half trillion floating point operations per second (TFLOPS) in under 10 watts or a full rack of NVIDIA T4 servers, delivering more than 10,000 trillion operations per second (TOPS) – DeepStream SDK provides a secure, reliable and scalable streaming framework for creating Intelligent Video Analytics (IVA) applications at the edge. The SDK provides an extensive set of features enabling developers and engineers to deliver real-time insights that are business critical.
With DeepStream, you can now quickly deploy IoT applications on edge devices such as Jetson Nano using Azure IoT Hub. The data from the edge device can then be sent to the cloud for further analysis. This makes it easier to unlock real-time insights and makes it easier and faster than ever before to detect and classify objects, recognize patterns and identify anomalies or defects.
Learn how to perform real-time analytics with DeepStream connected to Azure via Azure IoT edge. Get started by running a sample on GitHub and watch the demo in action on Microsoft Azure Internet of Things show.
If you are a business or a developer building embedded IoT applications, check out how DeepStream can remove bottlenecks without compromising on system performance and still have the ability to scale your next generation IoT solution with advanced Artificial Intelligence (AI).
DeepStream 4.0.1 is also available as a container image from NGC, registry for GPU-optimized deep learning frameworks, machine learning algorithms, and pre-trained AI models.
- DeepStream on Azure sample app on GitHub
- Watch DeepStream demo on Microsoft Azure Internet of Things show
- Access DeepStream from Azure marketplace
- If you are a developer looking to get started quickly, check out additional DeepStream SDK technical resources
Don’t miss out on all the latest NVIDIA developer news. Subscribe to our developer newsletter and be the first to get notified on new product announcements and updates.