Develop Multiplatform Computer Vision Solutions


Explore the OpenVINO™ toolkit (formerly the Intel® Computer Vision SDK)

Make your vision a reality on Intel® platforms—from smart cameras and video surveillance to robotics, transportation, and more.

Your Computer Vision Apps...Now Faster

Develop applications and solutions that emulate human vision with the Open Visual Inference & Neural Network Optimization (OpenVINO™) toolkit. Based on convolutional neural networks (CNN), the toolkit extends workloads across Intel® hardware and maximizes performance.

  • Enables CNN-based deep learning inference on the edge
  • Supports heterogeneous execution across computer vision accelerators—CPU, GPU, Intel® Movidius™ Neural Compute Stick, and FPGA—using a common API
  • Speeds time to market via a library of functions and preoptimized kernels
  • Includes optimized calls for OpenCV and OpenVX*

Get Started

Discover the Capabilities

Deep Learning for Computer Vision

Accelerate and deploy CNNs on Intel® platforms with the Intel® Deep Learning Deployment Toolkit that's available in the OpenVINO toolkit and as a stand-alone download.

Hardware Acceleration

Harness the performance of Intel®-based accelerators: CPUs, GPUs, FPGAs, VPUs, and IPUs.

Who Needs This Product

Software developers and data scientists who:

  • Work on computer vision, neural network inference, and deep learning deployment capabilities
  • Want to accelerate their solutions across multiple platforms, including CPU, GPU, VPU, and FPGA

What's New?

Deep Learning Enhancements

  • Open Neural Network Exchange (ONNX) support covers all public models on the ONNX Model Zoo.
  • The Computer Vision Algorithms (CVA) component includes three prebuilt algorithms: emotion recognition, person re-identification, and crossroad object detection.
  • Four pretrained models target smart classroom and re-identification use cases.
  • The Intel® Deep Learning Deployment Toolkit supports all object detection models on TensorFlow Model Zoo*. To achieve increased performance, deploy the models using the OpenVINO toolkit.
  • Model optimizer support is extended to include features of graph freeze and graph summary. It introduces general support for dynamic input freezing (via the command line), which helps to deploy models like FaceNet.
  • The inference engine now includes a preview of the image preprocessing capability (resize and crop) and static shape inferencing.

Traditional Computer Vision Updates

  • OpenCV 3.4.3 provides improved performance. This library features an initial Intel® Advanced Vector Extensions 2 support via universal intrinsics. It also supports GStreamer as a multimedia backend on Linux*.

Hardware Support

  • Multicard support for Intel® FPGAs. Boost deep learning performance even further by adding multiple FPGAs to your solution.

Recent Updates

Release Notes

Product Brief

What You Can Do

Security Camera Reference Implementation

Learn how to run a digital security and surveillance reference implementation with vehicle detection that includes vehicle attributes, such as license plate recognition.

What Customers Are Saying

“Intel® Xeon® Scalable processors and OpenVINO toolkit appears to be the right solution for medical imaging AI workloads. Our customers can use their existing hardware to its maximum potential, without having to complicate their infrastructure, while still aiming to achieve quality output resolution at exceptional speeds."

— Vijayananda J., chief architect and fellow, Data Science and AI, Philips HealthSuite Insights, India

White Paper

What Customers Are Saying

"Our strategic focus is bringing artificial intelligence on the edge. The challenge has been in implementing efficient algorithms to balance the workload between different hardware components. The OpenVINO toolkit bridges this gap, giving us a unique tool that speeds up the path to production, both for us and our partners."

— Fabrizio Del Maffeo, vice president and managing director, AAEON Technology, Europe

What Customers Are Saying

"Powered by the new OpenVINO toolkit, which delivers a comprehensive toolkit for developing and deploying [a] vision-oriented solution, and is capable of delivering higher computing and graphics performance, [several solution-ready packages from Advantech are launching] for vertical markets, including people counting, heat map analysis, facial recognition, and loss prevention. With the OpenVINO toolkit, we can easily process real-time video recording and transcoding, human detection and counting, as well as signage management."

— MC Chiang, associate vice president of Service-IoT, Advantech

What Customers Are Saying

"AI presents tremendous opportunity for industrial automation. With machine vision extensively used for guidance, identification, gauging, and inspection in manufacturing processes, the OpenVINO toolkit integrates deep learning tools, computer vision acceleration, and support for heterogeneous execution environments with high inference speed and accuracy for machine vision.”

— Edgar Chen, general manager of the embedded platforms and modules business unit, ADLINK*

What Customers Are Saying

"With the OpenVINO toolkit, we are now able to optimize inferencing across silicon from Intel, exceeding our throughput goals by almost six times. We want to not only keep deployment costs down for our customers, but also offer a flexible, high-performance solution for a new era of smarter medical imaging. Our partnership with Intel allows us to bring the power of AI to clinical diagnostic scanning and other healthcare workflows in a cost-effective manner.”

— David Chevalier, principal engineer, General Electric (GE) Healthcare*

What Customers Are Saying

"Dell EMC* surveillance labs validation program—and joint collaboration with Intel and its software tools—allows us to deliver proven solutions with the top surveillance camera and video software companies."

— Ken Mills, general manager, surveillance and security solutions for Dell EMC

Key Specifications

Not all OpenVINO toolkit features run on all combinations of processor systems. For more information, see System Requirements.

Development Platform

Processors:

6th to 8th generation Intel® Core™ and Intel® Xeon® processors

Operating systems:

  • Ubuntu* 16.04.3 LTS (64 bit)
  • CentOS* 7.4 (64 bit)
  • Windows® 10 (64 bit)

Target System Platforms

CPU:

  • 6th to 8th generation Intel Core and Intel Xeon processors
  • Intel® Pentium® processor N4200/5, N3350/5, N3450/5 with Intel® HD Graphics

Graphics:

  • 6th to 8th generation Intel Core processor with Iris® Pro graphics and Intel HD Graphics
  • 6th to 8th generation Intel Xeon processor with Iris Pro graphics and Intel HD Graphics (excluding the e5 product family, which does not have graphics)

FPGA & VPU:

  • Intel® Arria® 10 FPGA GX development kit
  • Intel Movidius Neural Compute Stick

IPU:

  • Intel® Atom™ processor E3900

Operating systems:

  • Yocto Project* version Poky Jethro v2.0.3 (64 bit)
  • Ubuntu* 16.04.3 LTS (64 bit)
  • CentOS* 7.4 (64 bit)
  • Windows® 10 (64 bit)