As movie-goers and video-gamers, we're accustomed to seeing life-like digital faces. Computer-generated countenances are also widely used in medicine, biometrics, and virtual reality. Still, it's very difficult to create faces that will fool us; human beings are wired, it seems, to pick up on very small differences in faces.
Prof. Hanspeter Pfister, Director of Visual Computing in the Initiative in Innovative Computing, has been working on this problem for years. In Analysis of Human Faces Using a Measurement-Based Skin Reflectance Model, Prof. Pfister and his colleagues describe some of the techniques they have developed for capturing and digitally reproducing an actor's performance and likeness.
You can find Prof. Pfister's works in DASH, many of which deal with digital representation of faces and places, here.