I am making photos to train cnn. So should I remove my hand with Photoshop. Hand is holding the product that should be detected. Will the training be ruined if I won't? Because there are lots of photos to work on, then. Am new in image detection. Thanks in advance!
If all of your images contain your hand, then the model will learn to look for your hand in the image and correlate that it's predictions. So if you put any object in your hand it's likely to make a false prediction, and if you predict on an an object without your hand it's less likely to give you a confident prediction.
Your training images should look like the images that you will predict on after the model is trained. If you plan on holding objects in your hand, taking a picture of that, and using that image to predict on then I think it's fine as is. If not, train on images that will look like what you will be predicting on.