frame

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Sign In

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Alert: Beginning Tuesday, June 25th, we will be freezing this site and migrating the content and forums to our new home at https://forums.intel.com/s/topic/0TO0P000000PqZDWA0/intel-neural-compute-sticks. Check it out now!

Tensorflow Detection object with Raspberry pi + NCS2

Hi.

Thanks to this tutorial I was able to detection object with raspberry. (https://github.com/EdjeElectronics/TensorFlow-Object-Detection-on-the-Raspberry-Pi)
but now i have NCS2, i need to do something to be able to run on it.
I had read the tensorflow manual on NCS but it didn't match NCS2
Thank you.

Comments

  • 4 Comments sorted by Votes Date Added
  • Hi @thienuittc
    If you're using the NCS2, the software kit that you'll use is OpenVINO. The openVINO toolkit can be installed on the Raspberry Pi 3, and here are the instructions. At the end of the instructions, there is a sample python script for face detection with OpenCV and the pre-trained face detection model. There are other pre-trained models included with the OpenVINO toolkit that can be used with your project. Here is a helpful blog about how to create an Object Detection program using the NCS2, OpenVINO and a Raspberry Pi. Here is another blog that might be useful to you.

    Please let me know if you have any further questions.
    Best Regards,
    Sahira

  • Hi.
    Thank you for all.
    But I’m using tensorflow, so now what should I do to be able to use it on NCS2
    I will wait for your response.

  • Open VINO is what you would use to convert from a Tensorflow graph to one that is compatible with the NCS2, how did you train your model?

  • Hi @thienuittc

    If you're using a Tensorflow model, you're going to use OpenVINO to perform inference on the NCS2. There are a couple steps you have to take before you can deploy your model onto the NCS2, You'll use the model optimizer to ultimately convert your tensorflow .pb model (.xml & .bin files) to an Intermediate Representation (IR) version. You can find the full instructions here.
    Please let me know if you have any further questions.

    Best Regards,
    Sahira

Sign In or Register to comment.