I am currently trying to create a One-Shot network using the Siamese architecture for an object that isn't a face.

My problem is, in normal Face Recognition the detecting gadget (e.g. Smartphone) knows which image it should compare the face, currently trying to unlock the tool, to.

In my case, I don't know the object's identity so I would have to compare every other object in the database to it.

Is there a better, more efficient way of checking the identity of the object without comparing it to any other object. A normal classification doesn't work in my case, because new unknown objects can be added any time.

  • $\begingroup$ You can store the embeddings of objects and just compare the embeddings, or store multiple and hen use something like kNN $\endgroup$
    – SajanGohil
    Mar 19, 2020 at 10:13


You must log in to answer this question.