Running Object Identification with Python and Webcam on a PC

[Kurdi]

Hello Edge Impulse team,
I am a beginner and I have a question regarding object identification using Python and a webcam (Iriun) without using a Raspberry Pi or Arduino.

I have seen that it is possible to run Edge Impulse on the web and test object identification using a webcam. I would like to achieve something similar but on my computer using Python. Specifically, I want to be able to run object identification and have the program print out the coordinates and the identified objects.

Is this possible with Edge Impulse? If so, could you provide some guidance or resources on how to do this?

Thank you for your help.

[shawn_edgeimpulse]

Hi @Kurdi,

Yes, you can train an object detection model using Edge Impulse and use it with Python to identify objects from a webcam. The following guides should offer a useful starting point:

• End-to-end detection project using MobileNet v2: https://docs.edgeimpulse.com/docs/tutorials/end-to-end-tutorials/object-detection/object-detection
• Edge Impulse supported object detection models: Object detection (Images) | Edge Impulse Documentation
• NVIDIA TAO models (enterprise only): NVIDIA TAO (Object detection & Images) | Edge Impulse Documentation
• Using the Linux Python SDK to perform inference: Linux Python SDK | Edge Impulse Documentation
• Example to capture an image from a webcam and perform inference (classification or object detection): linux-sdk-python/examples/image/classify.py at master · edgeimpulse/linux-sdk-python · GitHub