Hello how are you, I hope you are very well. This year we want to implement object detection in our robot. As we are new to this, we don’t know much about it and I wanted to know if you have any resources that can teach us about this and how we could implement it in our robot.
Hello,
Dataset Colab is a great resource to get started with. It offers a variety of pre-trained object detection models that you simply need to download and run on your robot. Additionally, there is basic documentation available to guide you through the process of running these models on your robot. I would be happy to answer any other more specific questions you might have!
Hope this helps,
Sean
2 Likes
If you have a limelight it has a super simple way to get started linked here Getting started with neural networks. We were able to set ours up in just a few minutes yesterday. (Note you will need a google coral)
3 Likes
Ohhh ok, thank you so much. I have another question. How can I add those files into the limelight using photonvision OS?
dataset colab is really designed for a custom vision system, completely designed in-house. the datasets from their website are just that- a large amount of images and annotations.
a better resource for you is you are new to this is what @ck2928 was saying about limelights. google corals are purchasable from andymark and that might be a better solution for you to spend time on. however, you will have to switch back to using limelight software, not photonvision.
photonvision does not natively support object detection, but i think there is someone on cd working on a branch that does photonvision supports object detection when used on an orange pi 5
And I will say it works quite well on the pi 5 right out of the box. Took us just a few minutes to get it going.