Page 36 - Geometric Modeling and Algebraic Geometry
P. 36

2 Some Covariants Related to Steiner Surfaces  33
                           2.6, presents the application of these covariants to the discrimination of classes of
                           parameterizations.


                           2.2 Orbits of quadratic parameterizations of quartics

                           A quadratic rational map from RP to RP is determined by a homogeneous
                                                                3
                                                         2
                           quadratic map f from R to R , that can be presented as a family of four real ternary
                                                   4
                                              3
                           quadratic forms:
                                f =(f 0 (x 0 ,x 1 ,x 2 ),f 1 (x 0 ,x 1 ,x 2 ),f 2 (x 0 ,x 1 ,x 2 ),f 3 (x 0 ,x 1 ,x 2 )) .  (2.3)
                           Denote with F the space of all the quadruples of real ternary quadratic forms. Then,
                           more precisely, quadratic rational maps from RP to RP can be identified with
                                                                    2
                                                                          3
                           the elements of F considered modulo scalar multiplication, i.e. the projective space
                           P(F).For f ∈F, we will denote with [f] the corresponding element of P(F).
                              Now the group GL(3, R) acts naturally on R (and RP ), and thus on F (and
                                                                           2
                                                                   3
                           P(F)). The action on F is as follows: for θ ∈ GL(3, R),
                                                      θ(f)= f ◦ θ  −1 .                   (2.4)
                           The induced action on P(F) corresponds to linear reparameterizations. There is also
                           a natural action of the group GL(4, R) on R (and RP ), and thus on F (and P(F)):
                                                                      3
                                                              4
                           for ρ ∈ GL(4, R),
                                                       ρ(f)= ρ ◦ f.                       (2.5)
                           We have thus an action of GL(3, R) × GL(4, R) on F (and P(F)). In the sequel, we
                           will denote this group with G.
                              In P(F), the subset U of those projective parameterizations with the property that
                                                      4
                           the Zariski closure of their image is a surface of degree 4 exactly, is invariant under
                           G. It is also a Zariski dense open set. As said in the introduction, the decomposition
                                                5
                           of U into orbits is known ; see [2, 7] and [8]. There are only six orbits. Table 2.1
                           provides the list of the orbits, with a representative for each.
                              Let us say a word about the connection between this problem and the analogous
                           problem in the complex setting. Denote with F C the complexification of F: that
                           is the space of families of four complex quadratic forms. Then P(F C ) represents
                           the space of quadratic rational maps from the complex projective plane, CP to the
                                                                                        2
                           complex projective three–dimensional space, CP . Let U C be the subset of those
                                                                   3
                           parameterizations whose image is a quartic surface. Then U is the trace of U C on
                           P(F). This means that U = U C ∩ P(F).
                              Let G C = GL(3, C)×GL(4, C). This group acts naturally on F C and P(F C ), and
                           also on U C . The classification of the orbits of P(F) under G is obtained by refining
                           the classification of P(F C ) into orbits under G C (see [1] for a modern reference about

                             We consider the set–theoretical image, and rule out the cases when the Zariski closure of
                            4
                             the image is a double quadric (case 7 in Proposition 5 of [2]) or a plane counted four times.
                             The determination of the orbits outside U is a different problem. See the references in [7].
                            5
   31   32   33   34   35   36   37   38   39   40   41