At the Max-Planck-Institut für Informatik in Saarbrücken Germany, a group of researchers led by Dr. Volker Blanz have created a program that will have implications in the fields of filmmaking and pornography (yes, you did hear that right..) and to a lesser extent to artists like me interested in painting the human face in all of its glories...
Their program takes a 2 dimensional image of a human face and using certain best fit algorithms can create a 3 dimensional model of the same face with expressions added to the face...Just think, now you could have accurate computer rendered models in high definition of how Mona Lisa would have looked if she were smiling, frowning, laughing or crying. Extending this further into the future - even a time lapse videography of Mona Lisa walking down the street. I do not even want to think how the porn industry might want to manipulate technology like this to satisfy individual perverse fantasies... Interesting technology nevertheless.
- The Morphable Face Model captures the variations of 3D shape and texture that occur among human faces. It represents each face by a set of model coefficients, and generates new, natural-looking faces from any novel set of coefficients, which is useful in a wide range of applications in computer vision and computer graphics.
- In this framework, it is easy to control complex facial attributes, such as gender, attractiveness, body weight, or facial expressions. Attributes are automatically learned from a set of faces rated by the user, and can then be applied to classify and manipulate new faces.
- Given a single photograph of a face, we can estimate its 3D shape, its orientation in space and the illumination conditions in the scene. Starting from a rough estimate of size, orientation and illumination, our algorithm optimizes these parameters along with the face's internal shape and surface colour to find the best match to the input image. The face model extracted from the image can be rotated and manipulated in 3D.
- In this framework, it is easy to control complex facial attributes, such as gender, attractiveness, body weight, or facial expressions. Attributes are automatically learned from a set of faces rated by the user, and can then be applied to classify and manipulate new faces.
- Given a single photograph of a face, we can estimate its 3D shape, its orientation in space and the illumination conditions in the scene. Starting from a rough estimate of size, orientation and illumination, our algorithm optimizes these parameters along with the face's internal shape and surface colour to find the best match to the input image. The face model extracted from the image can be rotated and manipulated in 3D.
You can get to a very mathematical description of the process that he uses here or you can view an animated clip of his experiments on images of Audrey Hepburn here.. I liked the video a lot. I am still trudging through the paper...
Some other papers of interest...
3D Videorealistic Facial Animation by Tony Ezzat, with Prof. T. Poggio and V. Blanz
Probing the Visual Representation of Faces With Adaptation : A View From the Other Side of the Mean by Fang Jiang, Volker Blanz and Alice J. O’Toole
Face Identification by Fitting a 3D Morphable Model using Linear Shape and Texture Error Functions by Sami Romdhani, Volker Blanz and Thomas Vetter
1 comment:
This would be incredibly helpful for me as I rely heavily on photo reference for my work. It sounds terribly expensive though. Looks like I'll have to stick to asking clients for every picture of the subject they have.
Post a Comment