What Is Transferred From Novel to Film？ Some Criticism of Brian McFarlane＇s Adaptation Analysis Method

Brian McFarlane＇s adaptation analysis method, first described in Novel to Film （1996）, is one of the most elaborated ones, and is often used in some form by contemporary scholars and critics. McFarlane attempts to use Roland Barthes＇ （1990） narrative theory for the comparison of novels and films, In any narrative, Barthes distinguishes two main groups of functions： functions proper and indices. McFarlane＇s ultimate aim with this taxonomy is to determine what elements of a novel are subject to adaptation proper and which ones are transferred directly into film. He concludes that a certain type of functions proper is transferable, while indices are more broadly open to adaptation than to the directness of transfer. The author finds McFarlane＇s theory problematic With the help of a Henry James＇ novella and its film adaptation, the author will try to demonstrate that functions proper are far too complex to enable a simple transfer. The author＇s point is that what is actually transferred from novel to film is something rather similar to the narrative functions described by early 20th century formalist Vladimir Propp

Local Properties of Cardinal Interpolatory Function

In [1], a kind of periodic cardinal interpolatory function (PCIF) is constructed, and they also use it to construct periodic wavelets. But the localizationl of PCIF is not known yet. In this paper, we studv the local property of PCIF using two different methods. The results show that PCIF has very good localization. Some new properties of Bernoulli spline are also introduced.

Performance driven expression mapping based on segmented examples

We present a method that combines performance-driven method with segmented 3D blendshape models to animate a face. First we prepare key sample examples and corresponding key target examples. Next we segment the whole face into two regions, for each region we reduce dimensionality of source examples using PAC into abstract space which is defined by truncated PCA eigen- vectors. Then for each example we fix the cardinal base function, which can determine the weight of the target example. Finally, in the animation stage we compute the weight of each example for each frame and add the weighted displacement vectors of each re- gion on the general face model.

On Cardinal Wavelets with Compact Support and Their Duals

Some people try to construct an orthonormal wavelet such that the corresponding scaling function（t）has the cardinal property,i.e.（n）=δn0,since such wavelets have many good applications.Unfortunate]y it is impossible to do so,except for a trivial case.In this work,a family of non-orthogonal cardinal wavelets with compact support is constructed and their duals are investigated.