Suche
News
You are here
Medieninformatik
By learning common properties of object classes from datasets of examples, the sites/projects explore how class-specific information can help to solve problems in image synthesis (Computer Graphics), image analysis (Computer Vision) and 3D shape processing. The work is based on a vector space representation of 3D shapes (Morphable Model). In graphics, learning from examples is becoming more and more relevant as the development of new rendering technology supports the creation of increasingly complex virtual scenes. In Computer Vision, learning-based methods provide general-purpose solutions, yet still adapt to the empirical properties and intrinsic structures of the problem during training.
Medieninformatik
Prof. Dr. Volker Blanz
Email: blanz@informatik.uni-siegen.de
Webseite: http://mi.informatik.uni-siegen.de
3D Shape Reconstruction and Manipulation
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.
The Morphable Face Model is derived from a data set of 3D face models by automatically establishing point-to-point correspondence between the examples, and transforming their shapes and textures into a vector space representation. New faces and expressions can be modeled by forming linear combinations of the prototypes.
Face Animation
Based on a dataset of 3D scans of different facial expressions and mouth shapes, we have formed a vector space of 3D expressions. Difference vectors, such as the smile-vector shown above, can be added to new individual faces. Unlike physical models, we observe the appearance of expressions, rather than simulating the muscle forces and tissue properties that cause the surface deformations.