We present two watermarking schemes for 3D meshes : - watermarking with geometrical invariant for fragile watermarking towards authentication and integrity purposes - watermarking in the geometrical spectral domain towards robust watermarking / Nous présentons deux schémas de tatouage pour maillages surfaciques 3D : - tatouage fragile par invariants géométriques pour l'authentification et l'intégrité - tatouage robuste dans l'espace de la décomposition spectrale

Le domaine de cette thèse est dans la topologie quantique et son sujet est axé sur l'interaction en- tre la topologie de basse dimension et la théorie des représentations. Ma recherche concerne as- pects différents des invariants quantiques pour les entrelacs et les $3$-variétés, visant a créer des ponts entre les façons algébriques et topologiques de les définir. D'une part, une description al- gébrique et combinatoire pour un concept mathématique, crée l'opportunité de développer des outils de calcul. D'un autre côté, les descriptions topologiques et géométriques ouvrent des per- spectives vers des constructions qui mènent a une compréhension plus profonde et a des théories plus subtiles.Les polynômes de Jones coloriés sont des invariants quantiques d'entrelacs contruits en partant de la théorie des représentations de $U_q(sl(2))$. Le premier invariant de cette séquence est le polynôme de Jones original, qui peut-être caractérisé aussi par la théorie de l'écheveau. Bigelow et Lawrence ont décrit un modèle homologique pour le polynôme de Jones. Ils ont utilisé la représentation de Lawrence, qui est une représentation de groupe de tresses sur l'homologie des revêtements d'espaces de configurations dans le disque pointé, et la nature de l'écheveau de l'in- variant pour la preuve. Contrairement a ce cas, les autres polynômes de Jones coloriés ne peu- vent pas être définis facilement par la théorie de l'écheveau.Dans la premiere partie de cette thèse, nous donnons un modèle topologique pour les polynômes de Jones coloriés. Nous utilisons leur définition comme invariants quantiques et construisons des correspondants topologiques pas à pas. Nous observons d'abord que l'invariant peut être codé par des espaces dits de plus haut poids, puis utiliser un résultat de Kohno, qui identifie ces espaces avec des représentations de Lawrence. Nous prouvons que les polynômes de Jones coloriés peu- vent être obtenus comme une forme d'intersection géométrique gradués entre des classes d'ho- mologie dans certaines couvertures des espaces de configuration de points dans le disque pointé.Les deuxième et troisième parties sont orientées vers les applications de la théorie de la représen- tation des super groupes quantiques aux invariants quantiques. La deuxième partie est une col- laboration avec N. Geer, ou nous construisons des invariants quantiques pour $3$-variétés a par- tir des représentations de $U_q(sl(2|1))$. Turaev-Viro ont défini une méthode de type somme d'état qui donne des invariants de $3$-variétés a partir de $ U_q(sl (2)) $. Pour les super groupes quantiques, cela entraîne l'annulation des invariants. Plus tard, Geer-Pa- tureau-Turaev ont défini une méthode modifiée qui commence par une catégorie avec de bonnes propriétés et conduit à des invariants non-nulls. Notre stratégie consiste a construire une caté- gorie qui peut-être utilisée dans cette méthode modifiée. La troisième partie concerne l'étude des algèbre centralisatrices pour les représentations de $ U_q (sl (2 | 1)) $. Wagner et Marin conjec- turaient les dimensions d'une suite d'algèbres centralisatrices correspondant à la représentation simple standard de $U_q(sl(2|1))$. Nous prouvons cette conjecture en utilisant des techniques combinatoires. / The domain of this thesis is within quantum topology and its subject is focused towards the interaction between low dimensional topology and representation theory. My research con- cerns different aspects of quantum invariants for links and $3$-manifolds, aiming to create bridges between algebraic and topological ways of defining them. On one hand, an algebraic and combinatorial description for a mathematical concept, creates the opportunity to develop computational tools. On the other hand, topological and geometrical descriptions open per- spectives towards constructions that lead to a deeper understanding and more subtle theories.The coloured Jones polynomials are quantum link invariants constructed from the representa- tion theory of $U_q(sl(2))$. The first invariant of this sequence is the original Jones polyno- mial, which can be characterised also by skein theory. Bigelow and Lawrence described a homological model for the Jones polynomial. They used the Lawrence representation, which is a braid group representation on the homology of coverings of configuration spaces in the punctured disk, and the skein nature of the invariant for the proof. In contrast to this case, the other coloured Jones polynomials cannot be defined in an easy manner by skein theory.In the first part of this thesis, we give a topological model for the coloured Jones polynomi- als. We use their definition as quantum invariants and construct step by step topological cor- respondents. We first observe that the invariant can be encoded through so-called highest weight spaces and then use a result by Kohno, which identifies these spaces with Lawrence representations. We prove that the coloured Jones polynomials can be obtained as graded geometric intersection pairings between homology classes in certain coverings of the config- uration spaces of points in the punctured disk.The second and third parts are oriented towards applications of representation theory of super quantum groups to quantum invariants.The second part is a collaboration with N. Geer, where we construct quantum invariants for$3$-manifolds from representations of $U_q(sl(2|1))$. Turaev-Viro defined a state-sum type method that gives $3$-manifold invariants from $U_q(sl(2))$. For super quantum groups, this leads to vanishing invariants. Later on, Geer-Patureau-Turaev defined a modified method which starts with a category with good properties and leads to non-vanishing invariants. Our strategy is to construct a category that fits into the input of this modified method.The third part concerns the study of centralizer algebras for representations of $U_q(sl(2|1))$. Wagner and Marin conjectured the dimensions of a sequence of centralizer algebras corre- sponding to the simple standard $U_q(sl(2|1))$-representation. We prove this conjecture us- ing combinatorial techniques.

by Wai-kwan Fong. / Bibliography: leaves 37-39 / Thesis (M.Phil.) -- Chinese University of Hong Kong, 1983

Thesis. 1978. Ph.D.--Massachusetts Institute of Technology. Dept. of Mathematics. / MICROFICHE COPY AVAILABLE IN ARCHIVES AND SCIENCE. / Vita. / Bibliography: leaves 132-137. / Ph.D.

Hung Ling Yan Lincoln. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2004. / Includes bibliographical references (leaves 87-88). / Abstracts in English and Chinese. / Chapter 1 --- Introduction --- p.5 / Chapter 2 --- Preliminary --- p.10 / Chapter 2.1 --- Background in geometry --- p.10 / Chapter 2.2 --- Background in harmonic maps --- p.12 / Chapter 3 --- Lie Point Transformations and Symmetries --- p.16 / Chapter 3.1 --- Definition of symmetries --- p.16 / Chapter 3.2 --- Determine the Lie point symmetries of partial differential equations --- p.25 / Chapter 3.2.1 --- Second order differential equations --- p.26 / Chapter 4 --- Similarity Variables --- p.30 / Chapter 4.1 --- "Similarity variables and group-invariant, solutions" --- p.30 / Chapter 4.2 --- Reduction of number of variables of the partial differential equations --- p.34 / Chapter 4.2.1 --- Determine the similarity variables --- p.34 / Chapter 4.2.2 --- Procedure to reduce the number of variables of a system of partial differential equations --- p.36 / Chapter 5 --- Group Invariant Harmonic Maps --- p.38 / Chapter 5.1 --- Determine the Lie point symmetries of the harmonic map equations --- p.39 / Chapter 5.2 --- Reduction of harmonic map equations to ordinary differ- ential equations --- p.54 / Chapter 5.3 --- Solving the harmonic map system which has been reduced to ordinary differential equations --- p.62 / Chapter 5.3.1 --- Case 1 of Theorem 5.2.1 --- p.62 / Chapter 5.3.2 --- Case 2 of Theorem 5.2.1 --- p.66 / Chapter 5.3.3 --- Case 3 of Theorem 5.2.1 --- p.75 / Bibliography --- p.87

令X 為個帶有凱勒形式(Kähler form ω) 以及全純四形式( holomorphic four- form Ω )的四維緊致卡拉比丘空間(Calabi-Yau manifolds) 。在一些假設條件下，通過研究Donaldson- Thomas方程所決定的模空間，我們定義了四維Donaldson-Thomas不變量。我們也對四維局部卡拉比丘空間(local Calabi-Yau four-folds) 定義了四維Donaldson-Thomas 不變量，並且將之聯繫到三維Fano空間的Donaldson- Thomas 不變量。在一些情況下，我們還研究了DT/GW不變量對應。最后，我們在模空間光滑時計算了一些四維Donaldson- Thomas不變量。 / Let X be a complex four-dimensional compact Calabi-Yau manifold equipped with a Kahler form ω and a holomorphic four-form Ω. Under certain assumptions, we de ne Donaldson-Thomas type deformation invariants by studying the moduli space of the solutions of Donaldson-Thomas equations on the given Calabi-Yau manifold. We also study sheaves counting on local Calabi-Yau four-folds. We relate the sheaves countings over X = KY with the Donaldson- Thomas invariants for the associated compact three-fold Y . In some specialcases, we prove the DT/GW correspondence for X. Finally, we compute the Donaldson-Thomas invariants of certain Calabi-Yau four-folds when the moduli spaces are smooth. / Detailed summary in vernacular field only. / Cao, Yalong. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2013. / Includes bibliographical references (leaves 100-105). / Abstracts also in Chinese. / Chapter 1 --- Introduction --- p.6 / Chapter 2 --- The *4 operator --- p.18 / Chapter 2.1 --- The *4 operator for bundles --- p.18 / Chapter 2.2 --- The *4 operator for general coherent sheaves --- p.20 / Chapter 3 --- Local Kuranishi structure of DT₄ moduli spaces --- p.22 / Chapter 4 --- Compactification of DT₄ moduli spaces --- p.34 / Chapter 4.1 --- Stable bundles compactification of DT₄ moduli spaces --- p.34 / Chapter 4.2 --- Attempted general compactification of DT₄ moduli spaces --- p.36 / Chapter 5 --- Virtual cycle construction --- p.39 / Chapter 5.1 --- Virtual cycle construction for DT₄ moduli spaces --- p.40 / Chapter 5.2 --- Virtual cycle construction for generalized DT₄ moduli spaces --- p.48 / Chapter 6 --- DT4 invariants for compactly supported sheaves on local CY₄ --- p.52 / Chapter 6.1 --- The case of X = KY --- p.52 / Chapter 6.2 --- The case of X = T*S --- p.57 / Chapter 7 --- DT₄ invariants on toric CY₄ via localization --- p.66 / Chapter 8 --- Computational examples --- p.70 / Chapter 8.1 --- DT₄=GW correspondence in some special cases --- p.71 / Chapter 8.1.1 --- The case of Hol(X) = SU(4) --- p.72 / Chapter 8.1.2 --- The case of Hol(X) = Sp(2) --- p.77 / Chapter 8.2 --- Some remarks on cosection localizations for hyper-kähler four-folds --- p.79 / Chapter 8.3 --- Li-Qin's examples --- p.80 / Chapter 8.4 --- Moduli space of ideal sheaves of one point --- p.83 / Chapter 9 --- Appendix --- p.85 / Chapter 9.1 --- Local Kuranishi models of Mc° --- p.85 / Chapter 9.2 --- Some remarks on the orientability of the determinant line bundles on the (generalized) DT₄ moduli spaces --- p.87 / Chapter 9.3 --- Seidel-Thomas twists --- p.90 / Chapter 9.4 --- A quiver representation of Mc --- p.92

No description available.

Let G be a simple algebraic group. Labelled trivalent graphs called webs can be used to produce invariants in tensor products of minuscule representations. For each web, a configuration space of points in the affine Grassmannian is constructed. This configuration space gives a natural way of calculating the invariant vectors coming from webs. In the case of G = SL_3, non-elliptic webs yield a basis for the invariant spaces. The non-elliptic condition, which is equivalent to the condition that the dual diskoid of the web is CAT(0), is explained by the fact that affine buildings are CAT(0). In the case of G = SL_n, a sufficient condition for a set of webs to yield a basis is given. Using this condition and a generalization of a technique by Westbury, a basis is constructed for SL_n. Due to the geometric Satake correspondence there exists another natural basis of invariants, the Satake basis. This basis arises from the underlying geometry of the affine Grassmannian. There is an upper unitriangular change of basis from the basis constructed above to the Satake basis. An example is constructed showing that the Satake, web and dual canonical basis of the invariant space are all different. The natural action of rotation on tensor factors sends invariant space to invariant space. Since the rotation of web is still a web, the set of vectors coming from webs is fixed by this action. The Satake basis is also fixed, and an explicit geometric and combinatorial description of this action is developed.

Thèse de doctorat : Mathématiques : Mathématiques pures : Metz : 2006. / Thèse soutenue sur ensemble de travaux. Bibliogr. p. 101-102.

Reproduction de : Thèse de doctorat : Mathématiques : Lyon 1 : 2004. / Titre provenant de l'écran titre. 33 réf. bibliogr.