Pretrained models

[lawHo]

Hi. Posted this on wrong forum first so I'll try again in here:

I'm wondering how to use Pretrained models, like for example the head-pose-estimation-adas-0001 -model. There are model.yml and a .prototext files in the folder but how do I use them in an application? Can I get the xml and bin files out of these files somehow? What about if I'd like to convert the model into a caffe-model, is that possible?

[Peh_Intel · Accepted solution]

Hi lawHo,

Greetings to you.

To use Intel pre-trained model for the application, the model must convert into Intermediate Representation (IR) files which contains xml and bin files. You don’t have to convert the Intel pre-trained model into a caffe-model.

Before start, you have to initialize the OpenVINO environment first with this command line below:

"<path_to_openvino>\bin\setupvars.bat"

Direct to the directory with this command line below:

cd <path_to_openvino>\deployment_tools\tools\model_downloader

Download the model using downloader.py with this command line below:

python downloader.py --name head-pose-estimation-adas-0001

3 folders (FP16, FP16-INT8, FP32) with each folder contains its respective xml and bin files will be created in the directory below:

<path_to_openvino>\deployment_tools\tools\model_downloader\intel\head-pose-estimation-adas-0001

Note: The xml and bin files for Intel pre-trained models can obtained directly by using downloader.py only.

Regards,

Peh

[Peh_Intel]

Hi lawHo,

This thread will no longer be monitored since this issue has been resolved. If you need any additional information from Intel, please submit a new question.

Regards,

Peh