I have tried using AI kit for almost 3 mouths, and I do some small project with AI kit, so I want to creat an open source course for AI kit. And here I have already build a github repo:GitHub - Seeed-Projects/Tutorial-of-AI-Kit-with-Raspberry-Pi-From-Zero-to-Hero: This repository provides a comprehensive step-by-step guide to building AI projects using the Raspberry Pi AI Kit.
8 Likes
We have complete contributor guidance:Tutorial-of-AI-Kit-with-Raspberry-Pi-From-Zero-to-Hero/ContributorGuidelines.md at main · Seeed-Projects/Tutorial-of-AI-Kit-with-Raspberry-Pi-From-Zero-to-Hero · GitHub
6 Likes
We have already started producing content for the course, if anyone would like to contribute, we are always welcome.
5 Likes
Please give a star to support our course.
2 Likes
Now，we have github io pages:
5 Likes
We recently used an AI kit and a Raspberry Pi to create a demo for managing retail supermarkets. In this demo, the AI kit runs YOLOv8n to detect products on the shelves, while EfficientNet runs on the CPU to detect warehouse intrusions. Additionally, we used Node-RED to create a dashboard. If you’re interested, you can watch the video below:
we will put this demo on our course this week.
5 Likes
I think this is absolutely fantastic work. Very instructive. Could you perhaps explain for me if its the same thing I need to do if I want to train a model to recognize my face specifically, as you did with the oranges? I’m new to all of this and I am so interested.
In fact, they are different things. The models used for facial recognition and object detection are not the same. Furthermore, the official AI kit does not currently support facial detection on the Raspberry Pi. Here is the link:
If you want to start an AI project, I recommend beginning with object detection.
And if you have any question feel free to ask.
When you say that it doesn’t support facial detection on the raspberry pi, do you mean that the official examples do not and that there would need to be tinkering on my side done? Or do you mean that the AI chip wont even allow it? Because what I’ve been able to find on there is that there sure are models for face-recognition/detection.
The AI chip support face recognition, but software things can not run on RaspberryPi now.
Could you please elaborate on what you mean with software things?
Do you mean for the model-training and such? I’ve got a stronger computer that gets that done if that’s what you mean!
No, it is nothing about training a model, it is about to develop a model on raspberrypi.
You can see not pipeline about face recognition.
Hi @oktober.yildiz , what @jiahao.li is saying is there is no out-of-the-box example application for face recognition in our Raspberry Pi 5 examples repository.
We do have customer that created biometrical face recognition application based on Hailo. We also do have face detection models in our ModelZoo:
GitHub - Hailo Model Zoo - Hailo-8 Face Detection
and we do have face recognition examples in our TAPPAS repository.
GitHub - Tappas - Face Recognition
But they will require a little effort to get running on the Raspberry Pi.
Thanks for the clarification!
I’ve been struggling with understanding all this different syntax, TAPPAS, HailoRT, Pipelines, etc. I’ll get there I hope
I’ve made alot of effort getting face detection and recognition working on the pi. Im still stuck I can get detection to work but can’t figure out how to process the detected faces or even print to the console when a face is detected. I also can’t figure out how to save the faces. maybe someone has any advice?
Core Electronics makes a lot of nice’n easy instructions on Pi. https://www.youtube.com/watch?v=3TUlJrRJUeM&t=382s
this one is about facial recognition and had success with it it easily
1 Like