Camera/video tech is becoming ever more sophisticated to the point where we now have cameras that can recognize and track faces. So, how long will it be before we start seeing cameras that can completely change a person's face in-camera? Never going to happen you say? Tell that to Arturo Castro and Kyle McDonald.
"Faces is an interactive installation by Arturo Castro and Kyle McDonald, extending work on face substitution. The piece resembles a mirror where people get their face swapped. This installation takes advantage of the unique experience of slowly recognizing yourself as someone else playing also with the concept of the uncanny valley by showing a face that perfecly matches the visitors facial expresion but still has some lighting and texturing imperfections. This elicits everything from laughter, to surprise, or repulsion."
"This is a technical demo for face substitution technique. The application works in real time and it's developed using the opensource framework for creative coding openFrameworks: openFrameworks.cc
Most of the "magic" happens thanks to Jason Saragih's c++ library for face tracking web.mac.com/jsaragih/FaceTracker/FaceTracker.html. The face tracking library returns a mesh that matches the contour of the eyes, nose, mouth and other facial features.
That way the mesh obtained from a photo is matched to my own face in the video. Applying some color interpolation algorithms from Kevin Atkinson's image clone code: methodart.blogspot.com/ gives it the blending effect that can be seen in the final footage.I'm also using Kyle McDonald's ofxFaceTracker addon for openframeworks github.com/kylemcdonald/ofxFaceTracker which wraps Jason's library for easier use.
Kyle has uploaded another video giving a try at a different blending algorithm, which get rids of the artifacts and an even creepier look: vimeo.com/29348533"
via [ArturoCastro] [KyleMcDonald]