tumbledry

Facial Attractiveness Enhancement

Data-Driven Enhancement of Facial Attractiveness sounds a bit dull at first, but consider what that means: an automated software approach to actually making faces more attractive. I must provide a picture illustrating the results (originals on the top, computer-enhanced results on the bottom):

faces

A quick summary of how this is done actually makes a lot of sense:

The key component in our approach is an automatic facial attractiveness engine trained on datasets of faces with accompanying facial attractiveness ratings collected from groups of human raters. Given a new face, we extract a set of distances between a variety of facial feature locations, which define a point in a high-dimensional “face space”. We then search the face space for a nearby point with a higher predicted attractiveness rating. Once such a point is found, the corresponding facial distances are embedded in the plane and serve as a target to define a 2D warp field which maps the original facial features to their adjusted locations.

To my eyes, this looks like an automated approach to accomplishing the same thing that professional retouchers do to magazine photos. It starts to explain how actors & actresses can resemble but not really look like themselves on the cover of these mags. Apparently, a demonstration application will be issued by this team, so it may be interesting to try the program out on faces we know. (via Waxy)

2 comments left

Comments

Alexander Micek

Please click on the “faces” tag to see what else has been tagged with “faces.” I was curious myself, and glad I checked it out.

Dan McKeown

Hey, I am one of the faces. Just as a side note, there are boobs in the background of that picture. Enjoy.

Brief Notes Nearby