Metro AI Suite Software Development Kit#

Provides a comprehensive and modular toolkit for accelerated media processing and AI inference, designed to fast-track the development of visual AI solutions.

Overview#

Provides a comprehensive and modular toolkit for accelerated media processing and AI inference, designed to fast-track the development of visual AI solutions.

The SDK supports accelerated media processing and inferencing with Intel ingredients such as Intel® Deep Learning Streamer framework, Intel® Distribution of OpenVINO™ toolkit, OneVPL, and Libva (VAAPI). It also includes enhanced versions of OpenCV and FFMpeg to speed up your Edge AI solutions development. Configure your application end-to-end with flexible AI capacity and reference video analytic pipelines for fast development.

  • Programming Language: Python, C, C++

  • Available Software:

    • Intel® Deep Learning Streamer

    • Intel® Distribution of OpenVINO™ toolkit

    • OpenCV(w/ OneVPL)

    • Intel® FFmpeg

Example Use Cases#

  • Smart Cities: Implement real-time traffic monitoring and management to reduce congestion and improve urban mobility.

    • Example: Use the SDK to develop applications that analyze traffic patterns and optimize signal timings.

  • Safety and Security: Enhance public safety with intelligent surveillance systems that detect and respond to potential threats.

    • Example: Deploy video analytics microservices to monitor public spaces and alert authorities to suspicious activities.

  • Transportation: Improve logistics and fleet management with AI-driven insights and automation.

    • Example: Utilize the SDK to track vehicle locations, predict maintenance needs, and optimize routes.

Key Benefits#

  • Scalable Platform: Supports a range of Intel processors, ensuring flexibility and performance for various edge AI applications.

  • Open-Source Foundation: Built on a popular, open-source operating system, providing a robust and customizable development environment.

  • Validated Dev Containers: Tailored for microservices development, these containers simplify deployment and management of AI workloads.

  • Video Analytics Microservice: Includes solution-level blocks for video analytics, enabling rapid development of visual AI applications.

  • Optimized for Visual AI: Designed specifically for visual AI workloads at the edge, leveraging Intel’s hardware accelerators.

  • Hardware Acceleration: Utilize onboard accelerators such as Intel® Iris® Xe graphics, Intel® Arc™ graphics cards, or the neural processing unit (NPU) in Intel® Core™ Ultra Processors.

Supported Hardware Platforms#

  • Intel® Core™:

    • 11th, 12th, or 13th Generation Embedded processors

    • Core™ Ultra Processors

    • 12th, or 13th Generation Desktop processors with Intel® Arc™ A380 Graphics

  • Intel Atom®: X7000 Series and N-series processors

  • Intel® Xeon®:

    • 4th, or 5th Generation Scalable Processors

    • Optional Pairing: Intel® Arc™ A380 Graphics

Other Prerequisites#

  • Operating System:

    • Ubuntu* 24.04.1 Desktop (fresh installation) for Intel® Core™ and Intel® Core™ Ultra platforms

    • Ubuntu* 24.04.1 Server (fresh installation) for Intel® Xeon® platform

  • System Requirements:

    • At least 128 GB of disk space

    • At least 16 GB of memory

  • Internet access

  • Superuser access (sudo)

How It Works#

Metro AI Suite Software Development Kit reference architecture forms the base to create a complete video analytic system for lightweight edge devices. The core of the SDK is a suite of quick-start container. Solution builders can leverage them as development containers to prototype and develop their AI workloads. These containers provide a comprehensive set libraries and toolkits for edge video analytic, starting from video decode to inference to encode and display.

Figure 1. Software Stack Diagram

Figure 1. Software Stack Diagram

The core of Metro AI Suite Software Development Kit reference architecture is a suite of containers. The SDK container includes OpenVINO™, oneDNN, and GPU drivers for accelerated inferencing and media processing,

Container

Content

Metro AI SDK

Video Analytics base libraries (media processing libraries + OpenCV + FFMpeg + OpenVINO Developer Tools + DLStreamer)

Edge Video Analytics Microservice

DLStremaer Pipeline Server to deploy optimized media analytics pipelines.

Learn More#

  • Get started with the microservice using the Get Started Guide.

  • Follow step-by-step examples to become familiar with the core functionality of the microservice, in Tutorials.