This is a market leading plug-and-play face verification SDK and model that will allow you to differentiate and store unique anonymous identities of game players. Send frames captured from the player's available webcamera (desktop, laptop, tablet, mobile) into the model and retrieve a face embedding, which you can think of as the "unique math" of a person's face. With an additional demographic estimation model we will be releasing soon, you will be able to estimate the age and gender of the user as well. 

You can compare face embeddings to ascertain the confidence that a user seen today is one of your known users, or if it's a new user. Build a database of the embeddings and you can search across a stored set of unique users and track them individual. Now your game knows who is playing.

Models running inside the SDK are downloadable per the documentation in the technical details.  

Stay tuned for subsequent model releases including demographic estimation and real-time measurement of attention and emotion!