I have a series of ordered pdf pages which own to different documents. Let me give you an example:
Pages: 1 2 3 4 5 6
True Pages: 1 2 | 1 2 3 4
So I have like six ordered pages, two of which from document A, and the remaining from document B. I do not have documents labels so the grouping should be done in an unsupervised way.
Which could be a reasonable approach? Using only CNN to detect border pages shouldn't be enough to discern documents, so I was thinking to something like RNN->CNN or CNN->RNN but I don't know how it would practically work because it is the first time I don't use labels in my TF model.
Do you think it would be a reasonable idea?