0

I have a mugshot (a photo of a person's face) and a picture (an arbitrary image file, can be literally anything: another mugshot, a group photo, a cartoon, a poster, a document scan).

I want to determine whether the person whose mugshot I have appears in the picture (with the confidence score).

Is there an OTS software to do that?

Obviously, since "face recognition" is a "solved problem", there is software that does it, the question is more like "where do I get it" and "how well does it work".

Thanks.

  • 1
    Face recognition is not a solved problem, although you can find software that will attempt it, it is still a very active area of research. Assuming you are capable of putting together the necessary ML pipeline, you could try OpenFace: cmusatyalab.github.io/openface – Neil Slater Jun 21 '17 at 16:30
  • @NeilSlater - THANKS! this is precisely what I was looking for! (and yes, I know that this is "a very active area of research" - this is why I put "solved problem" in quotes. Could you please turn your comment into an answer so that I could accept it and close the question? – sds Jun 21 '17 at 16:33
  • Actually I have voted to close the question, as IMO it's not really about Data Science. But I thought I'd link the library in the comments, since I'd like to help – Neil Slater Jun 21 '17 at 16:35
  • I think face recognition is very much about DS, and we already have the software-recommendation tag. Where do you think this question would be on topic?! – sds Jun 21 '17 at 16:36
  • 1
    "Third, the existence of the software-recommendation tag makes my question on-topic by default" I disagree. The question still needs to focus on a data science issue, not just use a product of data science research. Perhaps explain what data science you hope to do with the software? Otherwise the question is similar to e.g. "Please recommend me a style transfer image plugin?" or "Please advise on best choice for text autocomplete library in my messaging app?" – Neil Slater Jun 21 '17 at 16:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.