Using an image as input?

General
1

I have an object detection model in .hef format but I want it to detect based on images? I tried doing,
python basic_pipelines/detection.py --labels-json resources/barcode-labels.json --hef-path resources/model.hef --input resources/image.jpg

But I get an error
Frame count: 1

End-of-stream
Error rewinding the video.

Is it possible to do this? And how would I see what the model predicted the object in the image as?

2

Welcome to the Hailo Community!

The HEF file is only a small part of a image pipeline. We have many examples and tutorials to help you.

Inside the Hailo AI Software suite you can start the tutorials by calling:

hailo tutorial

This will start a Jupyter notebook server with notebooks for each step of the model conversion.

You can also find examples here:

GitHub - Hailo Application Code Examples

and for Raspberry Pi here:

GitHub - Hailo Raspberry Pi 5 Examples

and our Tappas GStreamer Examples

GitHub - Tappas - Gstreamer - General

I believe this examples allows you to input images:

GitHub - Hailo Application Code Examples - Python - Object Detection

3

Hi @gmbluth
You can try our PySDK which enables such use cases with simple code. You can see User Guide 1 Hailo World: Running Your First Inference on a Hailo Device Using DeGirum PySDK for an example.