Hi. I’m David Shaw. And welcome to AI News. In this episode, we discover
how the Intel Movidius Myriad 2 technology is used
for specialized vision processing at the edge
and how this effort can solve latency challenges
in end-to-end deep learning applications. [MUSIC PLAYING] The process of training CNNs
is very compute-intensive and can be greatly
enhanced in the cloud. However, cloud communication
introduces latency issues, which may lead to lagging
inference performance in edge devices and mission-critical
applications. Intel Fellowship recipient David
Ojika and graduate research assistant Vahid
Daneshmand set out to resolve this problem
using specialized vision processors and distributed
computing architecture. They explored end-to-end
image analytics with the Intel Movidius
Myriad 2 vision processing unit, the industry’s first
always-on vision processor. Their method, conclusions,
and future work are all examined in the
article, Solving Latency Challenges in End-to-End
Deep Learning Applications. In this article, you’ll
find out the workarounds for visualization support
and learn some easier ways to deploy deep neural networks. The Intel Movidius
Myriad 2 VPU can achieve real-time
performance for CNN inference on embedded devices. Ojika and Daneshmand proposed
a software architecture that presents inference
as a web service, enabling a shared platform
for image analytics on embedded devices and
latency-sensitive applications. Check out the link provided to
learn more about the project, don’t forget to like
this video and subscribe, and we’ll see you next
week for more AI News.

Intel® Movidius™ Myriad™ 2 | AI News | Intel Software

One thought on “Intel® Movidius™ Myriad™ 2 | AI News | Intel Software

  • June 10, 2018 at 10:50 am

    TL;DR "read the article in the description box"
    Why did you even make this video? Honestly.


Leave a Reply

Your email address will not be published. Required fields are marked *