Openvino in CVAT - model zoo - HRNet

[Thomascloostermans]

Is it possible to run HRNet-W32 (model zoo in openvino) in CVAT and if so how? I see a lot of commands to be used however I have no idea whether they are required if I want to run it in CVAT?

I'm completely new to this platform so I would really appreciate some tips and tricks regarding my question.

[IntelSupport]

Hi Thomascloostermans,

Thanks for reaching out.

For your information, CVAT is now not a part of OpenVINO toolkit. It is part of OpenCV, and I suggest you post your question on the CVAT community for further support.

However, there are videos for details explanations of CVAT with OpenVINO Toolkit that might be a help to you:

Introduction to the Computer Vision Annotation Tool (CVAT) | Intel Software

Semi and Full Automatic Annotation | CVAT | Intel Software

CVAT Computer Vision Annotation Tool | OpenVINO™ toolkit | Ep. 51 | Intel Software

CVAT Auto Annotation | OpenVINO™ toolkit | Ep. 52 | Intel Software

Please be noted, the last two videos are outdated but the concept of the CVAT with OpenVINO might be a help for you.

On another note, higher-hrnet-w32-human-pose-estimation model is used in Human Pose Estimation demo. Since CVAT can be implemented with Image Classification, Object Detection, Semantic, and Instance Segmentation, etc., the model might work with CVAT. There is an example of Semi-automatic and Automatic Annotation that uses the OpenVINO model.

Regards,

Aznie

[IntelSupport]

Hi Thomascloostermans,

This thread will no longer be monitored since we have provided information. If you need any additional information from Intel, please submit a new question. 

Regards,

Aznie