I'm looking for studies, academic publications, papers, blog post, or anything that relates to use cases in which image recognition have been used with highly similar inputs.
When using the word "similar" in my research, the results are biased, and yields use case to differentiate similar images, but not using similar images as input.
The thing I would like to do is to find out which methods are used for that (I believe that would be neural networks, but I would like to have deeper insights about this).