In this thesis, we present a method of generating the collage of a face from different individual pieces taken from real faces. We describe the process of preparing the data and parameterizing the faces with the help of Meshmonk. The thesis also describes various methods for combining and smoothing the faces. These methods depend on the logical meaning that Meshmonk ascribes to each vertex in the face. The obtained results of each method are compared against each other. We experimentally determine which parameters are key to generating a good collage.
|