I have a project where I need someone who has experience with Raspberry Pi 5 and HD micro cameras as well as AI generation in training the AI to detect an object via the camera.
I need the PI to use AI and ML to identify an image in the real time video feed and turn a servo. I already have a partially trained model and I have 2 hours of video from which I need to have the PI trained further. This project would involve 1. Taking the video already available and further training the AI but additionally 2. Adding 2 buttons so that when the PI identifies an incorrect object, I can tell it that the identification was incorrect (both from the captured video as well as real time) and or when the PI identifies correctly, the second button would allow me to tell it yes, you identified correctly thus to train the AI further.
I will provide the model for the PI and 2 hours of video .
The deliverable would be in 2 parts:
1. I would expect to receive the adjusted design so that I can continue to train the AI using more video and the button system so I can inform the AI when it is correct and incorrect.
2. I would need a procedure document created to teach me how to do what I would need to do to support the loading and further tuning.
3. I would expect a trained model useable in real world applications once the project concludes.