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.

Slow FPS on Neural Compute Stick 2 (2)

Hyodo__Katsuya
Innovator
419 Views

Hello.

I have the same problem as the post below.

https://software.intel.com/en-us/forums/computer-vision/topic/800112

I converted my own "UNet" model from .pb to .bin.

https://github.com/PINTO0309/TensorflowLite-UNet/raw/master/model/semanticsegmentation_frozen_person...

$ python3 mo_tf.py \
--input_model semanticsegmentation_frozen_person_32.pb \
--output_dir FP16 \
--input input \
--output output/BiasAdd \
--data_type FP16 \
--batch 1 

And I measured inference time for each combination of devices.

OpenVINO with

"CPU" vs "Neural Compute Stick v1" vs "Neural Compute Stick v2"

Inference time is longer for MYRIAD than for CPU.

The reasoning time measurement results are organized below.

Inference_Time

(1) Is not "Shave Core" used up to the maximum?

(2) Is there a way to speed up?

Please help me.

0 Kudos
2 Replies
Hyodo__Katsuya
Innovator
419 Views
Joel_E_
Beginner
419 Views

Thank you Katsuya for sharing!

Regards,

Joel

Reply