2019 Embedded Vision Summit

Event is now over

Find all of the presentations and other assets from this event in our Video and Presentations Library. Checkout some of our upcoming events over here.

2019 Embedded Vision Summit Banner
May 20-23, 2019
Santa Clara Convention Center, California, USA
Presentations from this event are now online here.

Don't miss this year’s OpenVX Workshop at Embedded Vision Summit. Khronos will present a day-long hands-on workshop all about OpenVX cross-platform neural network acceleration API for embedded vision applications. We’ve developed a new curriculum so even if you attended in past years, this is a do-not-miss, jam-packed tutorial with new information on computer vision algorithms for feature tracking and neural networks mapped to the graph API. We’ll be doing a hands-on practice session that gives participants a chance to solve real computer vision problems using OpenVX with the folks who created the API. We’ll also be talking about the OpenVX roadmap and what’s to come. 

Hardware Acceleration for Machine Learning and Computer Vision through Khronos Open Standard APIs

Date: Thursday May 23, 9am–5pm
Location: Rooms 203 and 204, Santa Clara Convention Center (MAP)

Applications needing inferencing and vision acceleration can use Khronos APIs for acceleration on multi-CPUs, GPUs, DSPs, and special-purpose neural processor hardware. This workshop introduces attendees to key Khronos open standards, including OpenVX, NNEF, OpenCL and SYCL, and shows how they can be applied to their projects and products.


Start Topic
9:00 AM Khronos/OpenCL (Neil Trevett, NVIDIA) – 40 minutes
9:40 AM OpenCL/SYCL (Andrew Richards, Codeplay) – 20 minutes
10:00 AM Break – 15 minutes
10:15 AM SYCL presentation (Konstantin S. Bobrovsky, Intel) – 45 minutes
11:00 AM OpenVX presentations (Frank Brill, Cadence / Niclas Danielsson, Axis / Mikael Pendse, Axis – 1 hour) 
12:00 PM Lunch – 1 hour
1:00 PM OpenVX presentation (Mike Schmit, AMD) – 30 minutes
1:30 PM NNEF presentation (Gergely Debreczeni, AImotive) – 30 min
2:00 PM NNEF hands-on* (Gergely Debreczeni, AImotive) – 30 min
2:30 PM OpenVX hands-on part 1** (Rajy Rawther and Kiriti Nagesh Gowda, AMD) – 30 minutes
3:00 PM Break – 15 minutes
3:15 PM OpenVX hands-on part 2** (Rajy Rawther and Kiriti Nagesh Gowda, AMD) – 2 hours
5:00 PM Finish

* A laptop with Python 2.7+ or Python 3.5+ installed is required for the NNEF hands-on portion

** Laptop must be able to connect to a remote server:

Target audience

Machine learning and computer vision application developers and managers

What will attendees gain by attending?

  • Understanding of the landscape of available APIs and standards for machine learning, inferencing and vision processing – and which are most applicable to their applications
  • Key technical insights into how to use Khronos standards and APIs for hardware acceleration across diverse hardware architectures
  • How their companies can influence the development of Khronos standards that affect their business

Fee: $25

Register today

Use promo code: MEMBER19 at checkout for 15% off any All-Access ticket to the event.


  • An introduction to the key concepts of Khronos APIs and standards for machine learning and computer vision, which include OpenVX, OpenCL, NNEF and SYCL
  • An overview of OpenVX, including the neural-network extension for implementation of inferencing
  • An introduction to the Neural Network Exchange Format (NNEF) as a standardized representation for bringing trained networks into inferencing engines
  • An introduction to Modern ISO C++ Heterogeneous dispatch showing how machine learning code written in C++ can be efficiently offloaded to accelerators including FPGAs, GPUs and Neural processors with SYCL
  • An update on OpenCL and how it is being extended for deployment flexibility across diverse, heterogenous hardware architectures
  • Practical demonstrations and code examples based on real world computer vision solutions and implemented using Khronos APIs will be used to explain key concepts

Other Khronos Related Sessions

Embedded Vision Seminar: Navigating Intelligent Vision at the Edge

Date and Time: Thursday May 23, 9:00am–5:00pm
Check-in: Badge pickup with coffee and pastries starts at 8:00 am.
Workshop Sponsored by:

Join us to learn about the latest trends in artificial intelligence and computer vision, and how to use the latest embedded vision technologies to navigate your way from concept to successful silicon. This seminar provides a deep dive into deep learning, embedded vision, and standards-based programming for automotive, mobile, surveillance, and consumer applications.

APIs for Accelerating Vision and Inferencing: an Industry Overview of Options and Trade-offs

Date and Time: Wednesday, May 22, 10:45 AM - 11:15 AM
Neil Trevett
Summit Track:
Technical Insights II

The landscape of SDKs, APIs and file formats for accelerating inferencing and vision applications continues to evolve rapidly. Low-level compute APIs, such as OpenCL, Vulkan and CUDA are being used to accelerate inferencing engines such as OpenVX, CoreML, NNAPI and TensorRT, being fed by neural network file formats such as NNEF and ONNX. Some of these APIs, like OpenCV, are vision-specific, while others, like OpenCL, are general-purpose. Some engines, like CoreML and TensorRT, are supplier-specific, while others such as OpenVX, are open standards that any supplier can adopt. Which ones should you use for your project?

Portable Performance via the OpenVX Computer Vision Library: Case Studies

Date and Time: Wednesday, May 22, 1:35 PM - 2:05 PM
Frank Brill
Summit Track:
Technical Insights II

OpenVX is a state-of-the-art open API standard for accelerating applications using computer vision and machine learning. The API and its conformance tests enable applications to leverage highly specialized features of hardware platforms while still retaining portability of application code across a wide range of architectures. This talk will use concrete examples on real implementations to demonstrate the performance portability of OpenVX. Example applications written using OpenVX will be described that run on platforms developed by Cadence Design Systems, Texas Instruments, Advanced Micro Devices and Axis Communications. Benchmarks will demonstrate performance gains that would otherwise only be achievable via hardware-specific code optimizations. The talk will also provide an update on the new features of the latest version of the OpenVX API, including support for a cross-platform neural network inferencing engine standard using a combination of OpenVX and Khronos’ Neural Network Exchange Format (NNEF).

Khronos Member Sponsors, Exhibitors and Speakers

  • AImotive
  • AMD
  • Arm
  • Cadence
  • CEVA
  • Codeplay
  • Google
  • Imagination Technologies
  • Intel
  • Mediatek
  • Mentor
  • Renesas
  • Synopsis
  • Texas Instruments
  • VeriSilicon

Conference Code of Conduct: The Khronos Group is dedicated to providing a harassment-free conference experience for everyone. Visit our Code of Conduct page to learn more.