Mentor(s) needed for object detection. No Idea what i am doing!

Hi all. Recovering developer (Retired) looking for something to do. I would like to work with YOLOv5 and need a lot of handholding. Been a few years since i did any development and it was mostly in the commercial investment space building data warehouses and general batch processing. I have an environment set up on my PC that works. Yolov5/Visual studio/python/pyTorch/Cuda and maybe more.

My next step is to learn how to train the system to recognize specific objects i am interested in. Any great tutorials for doing that? I am using some YouTube videos by Ivan Goncharov which have gotten me started, but i might need something more interactive.

Thanks for any ideas or help!

Hi @user102826 and welcome to the NVIDIA developer forums!

Unless you use out of the box solutions for object detection there is quite a bit to learn before you can run for example some Deep Learning Inference system on you home PC. A good place to start looking for self-paced or guided courses is NVIDIA’s Deep Learning Institute. There you will find content for every Deep Learning topic and lot of great learning resources to get you started.

Also Coursera is a great portal to learn fundamentals of AI, some of the courses being free (if you don’t want a certificate), and especially those by Andrew Ng can be recommended.

If you don’t want to do the learning part but instead jump right into some example code NVIDIA provides, I would suggest checking out one of our (slightly older, but still valid) Blog posts on Object detection in Python. Don’t worry that this is on Jetson, since it is based on Python you should be able to use this on PC as well with some adjustments. The “Hello AI” github is in general a good source for simple AI examples.

I hope this helps!