Openvino Segmentation Inferencing

[jeremyang012]

I have a question. I have been using the Training and Learning Suite AI Deep Learning (TLS) from Intel to do segmentation on the metal surface defect part. With that, I got the IR model (.xml and .bin) and want to use Python from the Pycharm platform to do segmentation inferencing. However, there is so far no proper Python coding for Window usage to integrate with this IR model. Most of the other codings are only used in the classification algorithms. Does anyone know any website that gives Python coding Window version to run the AI segmentation algorithm without using any Caffe model to convert?

[Megat_Intel]

Hi jeremyang012,

Thanks for reaching out to us.

To use your IR model for segmentation inference, you can try our tutorial Hello Image Segmentation on introduction to using segmentation models with OpenVINO™. You can also refer to this tutorial as well: Instance segmentation with OpenCV using Python.

Apart from that, check out our Open Model Zoo Demo, Image Segmentation Python* Demo, which does inference using semantic segmentation networks. Please note that only a few models are validated for this demo. You can use this demo as reference for your inference.

Additionally, I have searched for repositories with Python code for image segmentation on metal surface defects. I found several repositories and I share them here.

• Steel-Surface-Defect-Detection
• steel-defect-detection
• Metal-Surface-Defects-Detection-with-DL
• Metal-Surface-Defect-Inspection

Regards,

Megat

[Hairul_Intel]

Hi jeremyang012,

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

Regards,

Hairul