End-To-End Video Analytics – The Essential Tools & Techniques

Incorporating deep learning capabilities on a distributed heterogeneous system is hard enough. But add computer vision and media acceleration to the mix, and development challenges increase substantially.

This webinar focuses on using the OpenVINO™ toolkit to meet these types of challenges. Built to extend workloads across Intel® hardware and maximize performance on vision systems, OpenVINO can be used to deploy deep learning inference, implement computer vision, and accelerate media (decode/encode and image processing).

What you’ll learn:

  • A pragmatic approach to building complex workloads that run simultaneously across heterogeneous hardware, including CPU, GPU, FPGA, and Intel® Movidius™ VPU
  • How Intel® software tools work together to optimize your vision applications—OpenVINO, Intel® Media SDK, Intel® System Studio, and Intel® VTune™ Amplifier
  • Deep Learning advancements that enable pretrained models across a broader range of objects
    Coding the full workload for CPU (OpenCV*)
  • Intel-developed video analytics models optimized for specific hardware

OpenVINO is a trademark of Intel Corporation or its subsidiaries in the U.S. and/or other countries.

Guy Tamir, Video and Retail Software Solutions Manager, Intel Corporation

Guy Tamir has been with Intel since 1996, and is in the Intel® Software and Services Group (SSG). He has vast experience in both hardware and software design. He led engineering groups to design Intel® Core™ processor products, most recently as the integration manager for 6th gen Intel Core processors family (14nm). Guy was the product manager of the Intel® Computer Vision SDK (now known as OpenVINO™). Guy holds an MBA and M.Sc. Degree in Computer Engineering from Technion – Israel Institute of Technology.

For more complete information about compiler optimizations, see our Optimization Notice.