To train a new face on the platform, go to the Thesaurus tab.
You will directly arrive to the People page in which you will be able to drag and drop pictures of a person to teach them to the platform.
Drag and drop at least one photo (3 are recommended) in the blue box available on the top of the page.
To have the best learning process as possible for the platform, please respect these rules:
The face in the photo must be 300x300 pixels. Talking about the face and not the full photo
The face must be as neutral as possible and must not be partially hidden (with sunglasses for example)
The face must be looking straight at the camera or a slight three-quarter. We have to at least see both cheekbones of the person.
Once you've dropped the photos, a window Who is it? will pop up.
You will see a Name and Description field. Enter the name of the person, and you will see an autocomplete appear based on Wikidata data. Select the autocomplete data for an optimal learning (the description will be filled automatically as well).
If the person is known, a name suggestion will appear, you will then be able to click on Next. If the autocomplete offers a name not matching the person, click on Someone else to access name and description.
That's it, you just taught the platform a new face!
You didn't find the answer you were looking for?
Feel free to contact our support team using our chatbot or via email support@momentslab.com