How can I add additional data?
I want to add more object detection, specifically education related data. How would i add that data?
Also do we need to train for every single thing we need?
Hi, there are a few ways to train your own data.
Your first question: Yes, unfortunately you need to have a training dataset of all the things you want to train for. Thousands of images of each item. But you train once for all those different classes.
There are some online sites where you can train the YOLOv8 model. Where you upload your data.
Or if you have good GPU, you may be able to train offline. Currently Unity doesn't offer it's own service to train models.
Thanks for reply @PBird ,
I do have a good system and can train if share you some details on the process.
However my requirement is I need to detect text and image both but I am not sure if this model can do the job. Also phone camera will be used and not the video file.
It would be good if you can share some detail on possible options in unity.
Hi
@ryg81
. A better place to ask this would be in the Sentis discussions forum.
You are right this model is not good at reading text. You may be able to find the text but to read it you would need an OCR model.