NVIDIA is taking a significant leap forward with its latest expansion of the NVIDIA Metropolis Microservices. Now compatible with the NVIDIA Jetson embedded platform, this Cloud-based AI solution is set to revolutionize video streaming and AI-based perception applications at the edge.
Revolutionizing Edge AI with Jetson
The Metropolis Microservices on NVIDIA Jetson platform marks a pivotal advancement in creating cloud-native, secure vision AI applications. With the introduction of new APIs and microservices, NVIDIA is offering a path to modernize AI application stacks, streamline development processes, and ensure future-proof applications.
Empowering Developers with Flexible Microservices
NVIDIA’s approach involves packaging Metropolis Microservices as Docker containers, enabling developers to build modular and easily deployable systems. This flexibility allows developers to select specific microservices needed for their solutions, enhancing efficiency and customization.
Enhanced Video and AI Capabilities
The extension for Jetson includes support for video storage, media management, and advanced tracking algorithms. It also provides system monitoring and IoT services for seamless edge-to-cloud connectivity. Secure access, login, authentication, and analytics microservices further enrich the platform’s capabilities.
- Video Storage Toolkit: Offers hardware-accelerated video decoding and encoding, supporting H264 and H265 streaming protocols like RTSP and WebRTC.
- Analytics Service: Facilitates real-time tracking of objects in live or recorded video, leveraging NVIDIA’s DeepStream SDK for comprehensive video, audio, and image understanding.
Showcasing Solutions with Reference Applications
NVIDIA has introduced two pre-built reference applications to demonstrate the power of Metropolis Microservices: an AI-powered Network Video Recorder (AI-NVR) and a zero-shot detection application using generative AI. The AI-NVR showcases a complete pipeline from video ingestion to AI inference, analytics, and monitoring, complemented by a client app for Android. Meanwhile, the zero-shot detection application highlights the capabilities in generative AI applications running at the edge, controlled remotely via a REST API.
The Power of NVIDIA Jetson
The NVIDIA Jetson platform, designed for Edge AI and robotics applications, can perform up to 275 trillion operations per second (TOPS). It supports numerous open-source models optimized for the platform, available in the Jetson Generative AI Labs, paving the way for innovative edge computing solutions.
NVIDIA’s introduction of Metropolis Microservices for Jetson represents a bold step towards enabling developers to harness the power of AI at the edge, combining the robustness of the Jetson platform with the flexibility and scalability of cloud-native services.
Experts in development, customization, release and production support of mobile and desktop applications and games. Offering a well-balanced blend of technology skills, domain knowledge, hands-on experience, effective methodology, and passion for IT.
