The facial expression clone project involves modelling facial models with expressions for particular individuals. Creating a facial model from scratch is a labour intensive work, which involves several modelling pipelines; and crafting natural life-like facial expressions is really an artistic endeavour. So it would be useful to reuse existing data to automatically generate facial expressions on individual facial models. However, the facial models, created by different modelling tools or 3D scanners, might have different topological structure, shape and scale, which make a challenge to automatically find correspondences between each other and copy expressions. In this talk, I will present an automatic approach to transfer facial expressions from one facial model to another, which consists of three steps: automatic landmark matching, dense correspondence establishment, and facial expression transfer. The ICP (iterative closest point) partial matching algorithm is used to detect landmarks. The dense correspondences are built through iteratively solving an optimization problem, which minimizes the deformation energy of stretch and bend, subject to landmark constraints. The facial expression transfer is an optimization problem of minimizes deformation gradient. A demo will be present at the end.