Computer Vision Hardware
Choose the hardware accelerator that maximizes the performance of your application for any type of processor.
Intel® Processors
These CPUs offer the most universal option for computer vision tasks. With multiple product lines to choose from, you can find a range of price and performance options to meet your application and budget needs.
Common uses:
- Data analytics and cloud computing
- Energy efficient servers
- Network video recorders
Works best for:
- General-purpose processing, emphasizing performance and flexibility
- Real-world testing
- Quick-turn solutions (fast time to market)
- Solutions that have power or heat constraints
Resources:
Supported hardware:
- 6th to 10th generation Intel® Core™ processors and Intel® Xeon® processors
- Intel® Xeon® processor E family (formerly code named Sandy Bridge, Ivy Bridge, Haswell, and Broadwell)
- Intel® Xeon® Scalable processor (formerly Skylake and Cascade Lake)
- Intel Atom® processor with support for Intel® Streaming SIMD Extensions 4.1 (Intel® SSE4.1)
- Intel Pentium® processor N4200/5, N3350/5, or N3450/5 with Intel® HD Graphics
Supported operating systems:
- Windows® 10 (64 bit)
- Ubuntu* 18.04.3 LTS (64 bit)
- CentOS* 7.4 (64 bit)
- Yocto Project* version Poky Jethro 2.0.3 (64 bit)
- macOS* (64 bit)
Supported Intel® Distribution of OpenVINO™ toolkit components:
Intel® Processor Graphics
Many Intel processors contain integrated graphics, including Intel HD Graphics and Intel® UHD Graphics. The GPUs have a range of general-use and fixed-function capabilities (including Intel® Quick Sync Video) that can be used to accelerate media, inference, and general computer vision operations.
Common uses:
- Video encode, decode, and frame processing
- Deep learning inference offload
Works best for:
- Power-sensitive solutions
- Computer vision at the edge
Resources:
- Intel Quick Sync Video
- Overview of Intel Processor Graphics
- Intel® SDK for OpenCL™ Applications
- Supported Neural Network Layers
Supported hardware:
- 6th to 10th generation Intel Core processors with Intel® Iris® Pro graphics and Intel HD Graphics
- Intel Xeon processor with Intel Iris Pro graphics and Intel HD Graphics (excluding the e5 family which does not include graphics)
Supported operating systems:
- Windows 10 (64 bit)
- Ubuntu 18.04.3 LTS (64 bit)
- CentOS 7.4 (64 bit)
- Yocto Project version Poky Jethro 2.0.3 (64 bit)
Supported Intel Distribution of OpenVINO toolkit components:
Intel® FPGAs
Gain cost savings and revenue growth from integrated circuits that retrieve and classify data in real time. Use these accelerators for AI inferencing as a low-latency solution for safer and interactive experiences that can be applied to autonomous vehicles, robotics, IoT, and data centers.
Common uses:
- Real-time, low-latency inference
- FPGA abstraction using the Deep Learning Acceleration Suite
Works best for:
- Smart city
- Smart retail
- Smart factory
Resources:
Supported hardware:
- Intel® Arria® 10 FPGA GX Development Kit
- Intel® Programmable Acceleration Card with Intel® Arria® 10 GX FPGA
Supported operating systems:
- Ubuntu 18.04.2 LTS (64 bit)
- CentOS 7.4 (64 bit)
Supported Intel Distribution of OpenVINO toolkit component:
Common uses:
- Optimizing imaging, computer vision, and neural network pipelines
- Delivering high-performance, on-device deep learning inferences
- Furnishing data flow for machine intelligence workloads
- Supplying low power situations such as smart cameras or small compute devices
Works best for:
- Autonomous service robots and drones
- Digital security and surveillance
- Smart home and intelligent wearables
- Augmented reality (AR) and virtual reality (VR) all-in-one headsets
Resource:
Supported hardware:
Supported operating systems:
- Ubuntu 18.04.3 LTS (64 bit)
- CentOS 7.4 (64 bit)
- Windows 10 (64 bit)
- Raspbian* (target only)
- macOS (64 bit)
Supported Intel Distribution of OpenVINO toolkit components:
- Intel Deep Learning Deployment Toolkit
- OpenCV functions (need to be run against the CPU or GPU)
- Deep Learning Workbench
Intel® Vision Accelerator Design
FPGA
Available in a small form factor (as a PCIe* add-in card), this design enables deep learning inference at low power and low latency. It is well suited for real-time applications with limited space and power budget such as surveillance, retail, medical, and machine vision.
Common uses:
- Achieve optimized solutions through reprogramming flexibility
- Deliver high-performance, on-device deep learning inferences at low power and low latency
- Execute a compact form factor that's suitable for edge systems
Works best for:
- Industrial manufacturing
- Digital security and surveillance
- Retail
- Medical
Resource:
Supported hardware:
- Intel Vision Accelerator Design with an Intel Arria 10 FPGA (preview)
Supported operating system:
- Ubuntu 18.04.2 LTS (64 bit)
Supported toolkit component:
VPU
This design clusters multiple Intel® Movidius™ Vision Processing Units (VPU) (1~N) on an add-on card or rack-mount module server to provide deep learning inference acceleration. This family of vision accelerator design products comes in multiple form factors to cater to a wide range of vertical use cases.
Common uses:
- Optimization of the video analysis pipeline that includes video codec, image processing, computer vision, and neural network inference
- Real-time, low-latency inference
- Offload capability for high-density, deep-learning inference workloads
Works best for:
- Data center
- Cloud server, edge server, and edge network video recorder (NVR) server for surveillance
Resources:
Supported hardware with corresponding operating systems:
- Intel Vision Accelerator Design with Intel Movidius Vision Processing Units (VPU):
- Ubuntu 18.04 long-term support (LTS), 64 bit (Linux Kernel 5.2 and lower)
- Windows® 10, 64 bit
- CentOS 7.4, 64 bit
- AI Edge Computing Board with Intel® Movidius™ Myriad™ X C0 VPU, MYDX x 1:
- Windows 10, 64 bit
Supported component:
Common use:
- Free up additional bandwidth by offloading imaging functions that are normally run on other targets (such as a CPU or GPU).
Works best for:
- Print imaging workloads
Resources:
- Image Processing from Intel Presents New Possibilities for IP Cameras
- Product Brief
- Technical Library
Supported hardware:
- Intel Atom processor E3900 series
Supported operating systems:
- Yocto Project version Poky Jethro v2.0.3 (64 bit)
Ready to Get Started?
Try out hardware powered by the Intel Distribution of OpenVINO toolkit remotely using the award-winning1 Intel® DevCloud for the Edge.
Note Intel DevCloud for the Edge is currently available for enterprise developers only. Use your corporate email to apply.
1Intel DevCloud for the Edge is the 2020 Vision Product of the Year in the Developer Tool category as awarded by the Edge AI and Vision Alliance.
Discover the Capabilities
Traditional Computer Vision
Develop and optimize classic computer vision applications built with the OpenCV library and other industry tools.
Deep Learning Inference
Accelerate and deploy neural network models across Intel® platforms with a built-in model optimizer for pretrained models and an inference engine runtime for hardware-specific acceleration.
Product and Performance Information
1
Intel's compilers may or may not optimize to the same degree for non-Intel microprocessors for optimizations that are not unique to Intel microprocessors. These optimizations include SSE2, SSE3, and SSSE3 instruction sets and other optimizations. Intel does not guarantee the availability, functionality, or effectiveness of any optimization on microprocessors not manufactured by Intel. Microprocessor-dependent optimizations in this product are intended for use with Intel microprocessors. Certain optimizations not specific to Intel microarchitecture are reserved for Intel microprocessors. Please refer to the applicable product User and Reference Guides for more information regarding the specific instruction sets covered by this notice.
Notice revision #20110804