Intel® Distribution of OpenVINO™ Toolkit
Community assistance about the Intel® Distribution of OpenVINO™ toolkit, OpenCV, and all aspects of computer vision-related on Intel® platforms.

Wrong inference results on NCS2

Kaufmann__Elia
Beginner
630 Views

Hi there

I own two Movidius sticks, a NCS1 and a NCS2. Both seem to work fine (the example applications show correct results). However, when I convert my own Tensorflow models using the model optimizer, I get wrong results using the NCS2 while getting correct results on the NCS1. 

What could be the reason for this?

To reproduce this issue, I created a small github repo containing the converted model as well as all code and commands used to train, freeze and convert the model: git@github.com:kelia/movidius_tensorflow.git

If you want to reproduce the issue, do the following steps:

1. Clone repo

2. Install dependencies (openvino, opencv, tensorflow, colored)

    "sudo pip3 install tensorflow opencv-python colored"

3. Plug in Movidius stick

4. Run "python3 compare_results.py"

The script prints the results of 4 inference variants: plain tensorflow, froze graph, openvino myriad and openvino cpu. All 4 results should match, but on my machine the result for myriad on the NCS2 is wrong.

Thanks a lot for any help!

Elia

0 Kudos
3 Replies
anchor__jiang
Beginner
630 Views

i have same problem. do you solve it ?

0 Kudos
anchor__jiang
Beginner
630 Views

i have same problem

same model , ncs2 result is different from cpu gpu & ncs1

do you solve the problem

0 Kudos
whatthisismyname
Beginner
630 Views

hmmm, i am waiting for my NCSv2 to migrade the code from NCSv1 please tell me that this is just something to fix very simply.

0 Kudos
Reply