Intel® Distribution of OpenVINO™ Toolkit
Community support and discussions about the Intel® Distribution of OpenVINO™ toolkit, OpenCV, and all things computer vision-related on Intel® platforms.
5536 Discussions

Tensorflow Yolo v3 project is running too slow

PSINH5
Beginner
243 Views

How to use all the cores available on my Intel machine for a Tensorflow Yolo v3 object detection project? I believe it is using only one core at a time. What we want is a much higher performance, something similar to this :- https://software.intel.com/en-us/articles/tips-to-improve-performance-for-popular-deep-learning-fram...

0 Kudos
1 Reply
nikos1
Valued Contributor I
243 Views

Hello Parth,

YOLO v3 runs a bit slower than other networks. What is your inference device? CPU. GPU, MYRIAD? Tiny YOLO will run much faster if accuracy is acceptable in your application.

> I believe it is using only one core at a time.

How do did you find out? Using a profiler? I am seeing all CPU cores and GPU EUs utilized when I run YOLOv3.

What is you OS, Windows or Linux?

Are you using the OpenVino model optimizer to convert YOLO v3 pb to IR? What version of the OpenVino SDK are you using?

Cheers,

Nikos

 

Reply