Hello,
I have a compiled b0_segformer model. I got its successful hef file. Now I want to start inference from it. Meaning I want to give it images and get back results. Are there any instructions on how to implement this process against the hef model?
Thanks
David
Hi @dudibs,
We have many examples. I would start with running our simple CLI tool (hailortcli run ...
) to measure what is the FPS.
After this, you will need to implement some sort of a pipeline, to read in images (from a file, RTSP, directory, USB, or mipi camera), so a short preprocessing, feed to the Hailo device, and run the post processing and visualization.
We have many examples on our GitHub - hailo-ai/hailo-rpi5-examples and GitHub - hailo-ai/Hailo-Application-Code-Examples
thanks Nadav.
I am writing the application for semantic segmentation from scratch to infer semantic segmentation hef model from python
can you please send me example for this kind of model? i.e. inferring hef semantic segmentation model from python.
Hi Dudi,
You can check in our model zoo. We have a post processing function foe segformer.