Tech’s discussion of this town is all about the iPhone X facial recognition feature which can scans your head and create the already infamous “animojis.”
However, what if you don’t have $999 to put money into the next-generation smartphone but still wish to play with the 3D faces?
AI experts from the University of Nottingham and Kingston University today allow you to create a plausible re-creation of your face using a selfie or an image of yourself.
Try it yourself here, at the online presentation of their paper entitled “Large Pose 3D Face Reconstruction from a Single Image through Immediate Volumetric CNN Regression.”
Here’s what happened when I fed it with one pic of my head:
Here’s my colleague Rachel. Not precise, but not bad either:
Aaron S. Jackson, Adrian Bulat, Vasileios Argyriou, and Georgios Tzimiropoulos, who worked on the research, admit that 3D face reconstruction is a problem of “extraordinary difficulty.”
As the current systems operate, multiple facial pictures from various angles are fed into the machine which then needs to address several challenges, such as “establishing dense correspondences across large facial poses, expressions, and non-uniform lighting,” researchers said.
To overcome these challenges, Jackson and others trained a neural system — known as Convolutional Neural Network (CNN) — on a big fat dataset of both 2D pictures and 3D facial models.
The outcome is that the CNN can rebuild the whole 3D facial geometry from a unseen 2D picture. Not just that, it may suspect the parts of the face.
Pretty extraordinary, is not it?
Read more: http://mashable.com/