Intelligent Vision & Video

Accelerate your vision processing from edge to cloud.

The ability to combine high-definition (HD) video, image processing, analytics, and connectivity has become an essential feature for creating next generation intelligent vision and video applications.

Computer vision software developers must use the appropriate development environment for the hardware architecture of choice to achieve the right combination of performance, cost, and power efficiency for the design’s use case. Use of artificial intelligence and deep learning technologies for video analytics have moved out of the laboratory and into commercial use for many end markets, such as security and surveillance, retail, and machine vision-based manufacturing.

Intel products enhances vision systems capabilities with heterogeneous camera-to-cloud inference and deep learning acceleration solutions using the following:

  • Silicon - Intel® FPGAs, IA CPUs, IA CPU with integrated graphics, and Intel® Movidius™ Vision Processing Units (VPUs)
  • Software and intellectual property (IP) - the Intel® OpenVINO™ toolkit which includes the Intel® FPGA Deep Learning Acceleration Suite

Intel® FPGAs and SoCs, along with IP cores, development platforms, and a software developer design flow, provide a rapid development path with the flexibility to adapt to evolving challenges and solutions in each part of the video or vision pipeline for a wide range of video and intelligent vision applications.

 

 

Applications

Features

Real-Time Analytics

Intelligent, real-time video analytics are key functions in many vision applications. Typically, multiple algorithms such as motion detection, facial recognition, and object detection are needed to run simultaneously and across multiple camera sources. With Intel FPGAs and the Intel FPGA Deep Learning Acceleration Suite, designers can take advantage of industry-standard artificial intelligence frameworks, models, and topologies to create FPGA-powered versatile accelerators to implement convolutional neural network inferencing engines. By offloading the computation-intensive workloads from the Host CPU to the FPGA accelerator, designers can optimize the implementation of these multiple algorithms across the CPU + FPGA resources to improve overall system performance.

The FPGA Advantage at Each Step of the Video / Vision Pipeline

Intel® FPGAs offer performance, cost, flexibility, and integration advantages at each step of the video or vision pipeline. 

Flexible Sensor Interfaces

Image sensor suppliers often have proprietary interfaces, and the sensor interfaces keep evolving to keep up with improved sensor capabilities. Intel FPGAs enable designers to adopt new sensor interfaces easily without changing the rest of the design. In contrast, ASSPs require designers to adopt a new ASSP and redesign the camera board.

High-Performance Video and Image Signal Processing

Intel® FPGAs enable designers to design high-performance video and signal processing systems quickly and flexibly. IP cores such as Intel’s Video and Image Processing Suite allow designers to pick and choose video processing functionalities as needed, while the high-performance FPGA fabric can be leveraged to accelerate video pre-processing of high-resolution videos.

Universal Connectivity & Bridging

The abundance of different connectivity standards in today’s video and vision systems make FPGAs the ideal platform for universal connectivity and bridging.

Intel and our IP partners offer a variety of connectivity options, such as DisplayPorthigh-definition multimedia interface (HDMI), and serial digital interface (SDI) IP cores, CoaXPress, Camera Link, GigE Vision, USB3.0, IEEE802.1x Time-Sensitive Networking (TSN) Ethernet, MIPI, and others.

High-Efficiency Video Compression

Highly-efficient compression standards such as H.264/H.265 allow the use of smaller FPGAs to reduce system cost.  Integration of video CODECs with other video processing functions on a single FPGA enables designers to achieve large bill of material (BOM) savings. Intel offers multiple video compression IP cores, which allow designers the flexibility to support multiple compression standards and profiles depending on the application requirements. Contact us to hear about one of the hottest trends in vision processing: cloud and network hosted digital media functionality.

Additional Resources

Type Application  Description
Article OpenCL™

OpenCL™ Streamlines FPGA Acceleration of Computer Vision

(Registration/Log-in to the Embedded Vision Alliance website required to access the entire article.)

For more information about embedded vision, including hundreds of additional videos, please visit the Embedded Vision Alliance website at http://www.embedded-vision.com.

 

Article OpenCL™

OpenCL Eases Development of Computer Vision Software for Heterogeneous Processors

(Registration/Log-in to the Embedded Vision Alliance website required to access the entire article.)

For more information about embedded vision, including hundreds of additional videos, please visit the Embedded Vision Alliance website at http://www.embedded-vision.com.

 

Video OpenCL™

OpenCL Optimization Techniques: Image Processing Algorithm Example

This training provides a simple overview of an architectural optimization approach for targeting OpenCL on an FPGA for image processing algorithms.

Article General

Sorting Out Embedded Vision Systems

Papers at this year’s Embedded Vision Summit suggested the vast range of ways that embedded systems can employ focused light as an input, and the even vaster range of algorithms and hardware implementations they require to render that input useful. Applications range from simple, static machine vision to classification and interpretation of real-time, multi-camera video.

Article Automotive

Smart Exterior Cameras Enhance Vehicle Safety, Security

(Registration/Log-in to the Embedded Vision Alliance website required to access the entire article.)

For more information about embedded vision, including hundreds of additional videos, please visit the Embedded Vision Alliance website at http://www.embedded-vision.com.

Article Automotive

Complex Trends and Challenges in Designing ADAS Systems

In the race to develop reliable and cost-effective advanced driver assistance systems (ADAS), designers are presented with challenges to integrate functionality, develop scalable platforms and design systems that are robust enough to work in various operating conditions.[…]

Video General

4K Video & Image Processing (VIP)

Hear about the challenges that 4K video processing brings, and how the Intel® VIP Suite solves these challenges.

Video General

Video Systems 101

This video walks through some of the basics of video imaging and processing algorithms, connectivity, chroma sampling, and color formats.

Video General

Streamline Your Video Processing Apps with Design Examples on the VEEK

This video demonstrates a number of video and image processing functions based on Intel's Video and Image Processing Suite.  The demonstration uses the Video & Embedded Evaluation Kit (VEEK) provided by Terasic.

Related Links:

VEEK on Wiki

Purchase the VEEK

Video and Image Processing Suite

  CNN

Object Detection and Recognition with Neural Networks

See how our partner iAbra demonstrates machine learning with convolutional neural networks on FPGAs using the OpenCL™ flow to accelerate object detection and recognition scenarios.

Video Military

High Resolution Multi Camera System for Situational Awareness

This video shows a demonstration of a situational awareness application using multiple feeds with a Stratix® V FPGA performing as the main aggregator of all the video data in the system.

 

Video General

UDX10 4K60 Video Processing Reference Design

4 minutes

This video shows a demonstration of up/down/cross conversion 4K60 UHD (RGB 4:4:4 10b) reference design with HDMI 2.0 running on Intel® Arria® 10 FPGA.

Video General / Broadcast

Multichannel Full HD H.265 Codec Demonstration

This video shows a demonstration of a multichannel, 1080p60 H.265 CODEC running on a single FPGA, with ultra low latency.

Additional Resources

Need help with your FPGA design?

Collaborate with Intel on your next project.

Contact us

Intel® FPGA Topics

Learn how to leverage the power of Intel® FPGAs.

View all topics

Intel® FPGA and Programmable Devices

Learn how these powerful devices can be customized to accelerate key workloads and enable design engineers to adapt to emerging standards or changing requirements.

View all devices

Ürün ve Performans Bilgileri

1

OpenCL™ and the OpenCL logo are trademarks of Apple Inc. used by permission by Khronos.