Active shape models is a popular technique to object extraction. To this end, one models the geometric form of the object of interest. Object extraction is then equivalent with seeking a linear/non-linear transformation that projects/deforms this geometric form to an image region with the desired visual properties. The definition of the image term is the most challenging component of such an approach. Level set methods is a powerful optimization framework, that can be used to recover objects of interest by the propagation of curves. They can support complex topologies, considered in higher dimensions, are implicit, intrinsic and parameter free. Furthermore, one can introduce various image terms when seeking an object of particular form/visual properties. In this paper we re-visit active shape models and introduce a level set variant of them. Such an approach can account for prior shape knowledge quite efficiently as well as use data/image terms of various form and complexity while being able to deal with important local deformations and changes of topology. Promising experimental results demonstrate the potential of our approach.