Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Supported by

Newbie looking for help

Hi, I am an OpenSesame virgin, trying to use it for my undergraduate project. Someone told me that there are quite a few stock face images on OpenSesame, but I have spent ages looking and can't locate them. If anyone could help me, I would really appreciate it. Also any advice/tutorials on facial recognition would be fantastic. Many thanks.

Comments

  • Hi there.

    With facial recognition, you mean facial expression identification, or really identifying faces? Both topics are really hard to cover if you are just starting with OpenSesame, but I'll give you some leads anyway. You can check out https://www.affectiva.com/ for facial expression analysis, but as far as I know, they only provide an algorithm to do this and not an implemented solution. For facial location/identification, there are python libraries such as face_recognition, or MTCNN. You will need a very powerful graphical card to use these though. Or you can try out this library on Amazon

    Buy Me A Coffee

  • Hi, thank you so much for responding. I am attempting an experiment using facial recognition, but simple, just 24 faces with hair in the initial learning phase and then 48 faces, 24 of which are shown with hair, and 24 without in the recognition phase. Participants just have to press one key if they recognise the face and another if they don't. Do you know any tutorials that could aid me with this?

  • I also have faces and have already altered them so that the hair is erased on the 24 that are presented with no hair.

  • Hi,

    That is not too complicated.

    You need two loops, one for learning, one for memorizing. In the learning phase, you randomly select 24 of your faces and present them to participants in a way that you want. In the recognition phase, you add the pictures without the hair. The details depend on your specific issues. As a starting point you can use the beginners tutorial. To add the learning/recognition phases, I would use inline_scripts that randomize and assign the images to the right phase.

    Let me know if you need help.

    Eduard

    Buy Me A Coffee

Sign In or Register to comment.