1

for my thesis i need a face dataset where the same person is available at different ages, and possible in more real-life-like photos rather than portraits.

  • Data: i'm looking for a free to use dataset of faces with possibly multiple photos of the same person at different ages, and more real-life-like photos rather than portraits.
  • Context: showing how the face-recognition services i tested for my thesis (AWS, Azure and Google) aren't reliable enough to be integrated in a Smart Conversational Agent for Recognition.
  • License: free, with the possibility of showing 10-20ish of the pictures in my thesis Format: whether it's a database with a prebuilt table, a simple zip, or even a set of links to individual pictures, it doesn't matter. The latter (set of links) would actually be more helpful since the web service i made to interface myself with AWS-Azure-Google uses links, but it's not a requirement at all.
  • Authority: doesn't matter, i just want my degree ahahahah

The purpose of the project is testing AWS, Azure and Google's face recognition services to identify the same person across multiple photos (with one or more people) in a wide age difference, to eventually include facial recognition in a conversational agent for reminiscence. I already tested enough with random photos to know the result (that is none of these services is reliable enough to identify the same person with a wide age gap and with little training), but i need a dataset for which i can have the rights to use in my thesis.

Is there any online and freely available? I've checked some of the ones listed here but they all seem to have only portraits with no group photos nor any significant age gap for the same person.

  • Hello again. You can give this a look to improve your post. :) – pmcarpan Apr 20 at 16:38
  • that's deep, i'll take a look – Barnack Apr 20 at 16:42

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.