The technique of recursive subdivision can be visualised, loosely, as successively chopping off the corners of a polyhedron to make it less pointed. If the polyhedron is represented as a mesh of points connected by edges, repeated application of the subdivision results in progressively finer meshes tending in the limit to a surface. The subdivision is determined by the weightings given to the respective points and their neighbours.

No description available.

No description available.

Les clones de visage 3D peuvent être utilisés dans de nombreux domaines tels que l'interaction homme-machine et comme prétraitement dans des applications telles que l'analyse de l'émotion. Toutefois, ces clones doivent avoir la forme du visage bien modélisée tout en conservant les spécificités des individus et ils doivent être sémantiques. Un clone est sémantique quand on connaît la position des différentes parties du visage (yeux, nez ...). Dans notre technique, nous utilisons un capteur RVB-Z pour obtenir les spécificités des individus et un modèle déformable de visage 3D pour marquer la forme du visage. Pour la reconstruction de la forme, nous inversons le processus utilisé classiquement. En effet, nous réalisons d'abord le fitting puis la fusion de données. Pour chaque trame de profondeur, nous gardons les parties appropriées de données appelées patchs de forme. Selon le positionnement de ces patchs, nous fusionnons les données du capteur ou les données du modèle déformable de visage 3D. Pour la reconstruction de la texture, nous utilisons des patchs de forme et de texture pour préserver les caractéristiques de la personne. Ils sont détectés à partir des cartes de profondeur du capteur. Les tests que nous avons effectués montrent la robustesse et la précision de notre méthode. / 3D face clones can be used in many areas such as Human-Computer Interaction and as pretreatment in applications such as emotion analysis. However, such clones should have well-modeled facial shape while keeping the specificities of individuals and they should be semantic. A clone is semantic when we know the position of the different parts of the face (eyes, nose...). In our technique, we use a RGB-D sensor to get the specificities of individuals and 3D Morphable Face Model to mark facial shape. For the reconstruction of the shape, we reverse the process classically used. Indeed, we first perform fitting and then data fusion. For each depth frame, we keep the suitable parts of data called patches. Depending on the location, we merge either sensor data or 3D Morphable Face Model data. For the reconstruction of the texture, we use shape and texture patches to preserve the person's characteristics. They are detected using the depth frames of a RGB-D sensor. The tests we perform show the robustness and the accuracy of our method.

Les clones de visage 3D peuvent être utilisés dans de nombreux domaines tels que l'interaction homme-machine et comme prétraitement dans des applications telles que l'analyse de l'émotion. Toutefois, ces clones doivent avoir la forme du visage bien modélisée tout en conservant les spécificités des individus et ils doivent être sémantiques. Un clone est sémantique quand on connaît la position des différentes parties du visage (yeux, nez ...). Dans notre technique, nous utilisons un capteur RVB-Z pour obtenir les spécificités des individus et un modèle déformable de visage 3D pour marquer la forme du visage. Pour la reconstruction de la forme, nous inversons le processus utilisé classiquement. En effet, nous réalisons d'abord le fitting puis la fusion de données. Pour chaque trame de profondeur, nous gardons les parties appropriées de données appelées patchs de forme. Selon le positionnement de ces patchs, nous fusionnons les données du capteur ou les données du modèle déformable de visage 3D. Pour la reconstruction de la texture, nous utilisons des patchs de forme et de texture pour préserver les caractéristiques de la personne. Ils sont détectés à partir des cartes de profondeur du capteur. Les tests que nous avons effectués montrent la robustesse et la précision de notre méthode. / 3D face clones can be used in many areas such as Human-Computer Interaction and as pretreatment in applications such as emotion analysis. However, such clones should have well-modeled facial shape while keeping the specificities of individuals and they should be semantic. A clone is semantic when we know the position of the different parts of the face (eyes, nose...). In our technique, we use a RGB-D sensor to get the specificities of individuals and 3D Morphable Face Model to mark facial shape. For the reconstruction of the shape, we reverse the process classically used. Indeed, we first perform fitting and then data fusion. For each depth frame, we keep the suitable parts of data called patches. Depending on the location, we merge either sensor data or 3D Morphable Face Model data. For the reconstruction of the texture, we use shape and texture patches to preserve the person's characteristics. They are detected using the depth frames of a RGB-D sensor. The tests we perform show the robustness and the accuracy of our method.

International Telemetering Conference Proceedings / October 27-30, 1997 / Riviera Hotel and Convention Center, Las Vegas, Nevada / An edge-based hybrid finite element boundary integral (FE-BI) method using cylindrical shell elements is described for analyzing conformal quarter-wave patches embedded in a circular cylinder. Special care is also taken to deal with weight functions, dyadic Green’s function, and feed model. Some types of the patch arrays embedded in different circular radius have been developed. The tests of their VSWRs and radiation characteristics are in good agreement with the theoretical results.

When interpolating a data mesh using triangular Bézier patches, the requirement of C¹ or G¹ continuity imposes strict constraints on the control points of adjacent patches. However, fulfillment of these continuity constraints cannot guarantee that the resulting surfaces have good shape. This thesis presents an approach to constructing surfaces with approximate C¹/G¹ continuity, where a small amount of discontinuity is allowed between surface normals of adjacent patches. For all the schemes presented in this thesis, although the resulting surface has C¹/G¹ continuity at the data vertices, I only require approximate C¹/G¹ continuity along data triangle boundaries so as to lower the patch degree. For functional data, a cubic interpolating scheme with approximate C¹ continuity is presented. In this scheme, one cubic patch will be constructed for each data triangle and upper bounds are provided for the normal discontinuity across patch boundaries. For a triangular mesh of arbitrary topology, two interpolating parametric schemes are devised. For each data triangle, the first scheme performs a domain split and constructs three cubic micro-patches; the second scheme constructs one quintic patch for each data triangle. To reduce the normal discontinuity, neighboring patches across data triangle boundaries are adjusted to have identical normals at the middle point of the common boundary. The upper bounds for the normal discontinuity between two parametric patches are also derived for the resulting approximate G¹ surface. In most cases, the resulting surfaces with approximate continuity have the same level of visual smoothness and in some cases better shape quality.

When interpolating a data mesh using triangular Bézier patches, the requirement of C¹ or G¹ continuity imposes strict constraints on the control points of adjacent patches. However, fulfillment of these continuity constraints cannot guarantee that the resulting surfaces have good shape. This thesis presents an approach to constructing surfaces with approximate C¹/G¹ continuity, where a small amount of discontinuity is allowed between surface normals of adjacent patches. For all the schemes presented in this thesis, although the resulting surface has C¹/G¹ continuity at the data vertices, I only require approximate C¹/G¹ continuity along data triangle boundaries so as to lower the patch degree. For functional data, a cubic interpolating scheme with approximate C¹ continuity is presented. In this scheme, one cubic patch will be constructed for each data triangle and upper bounds are provided for the normal discontinuity across patch boundaries. For a triangular mesh of arbitrary topology, two interpolating parametric schemes are devised. For each data triangle, the first scheme performs a domain split and constructs three cubic micro-patches; the second scheme constructs one quintic patch for each data triangle. To reduce the normal discontinuity, neighboring patches across data triangle boundaries are adjusted to have identical normals at the middle point of the common boundary. The upper bounds for the normal discontinuity between two parametric patches are also derived for the resulting approximate G¹ surface. In most cases, the resulting surfaces with approximate continuity have the same level of visual smoothness and in some cases better shape quality.

This work is aimed to understand the mechanisms for evolution and reversed evolution of dislocation structure under variable strain amplitudes, using automotive-grade interstitial-free steels (IF steel) under strain ratio (R) = 0 condition. The microstructures were mainly examined by the SEM under BEI/ECCI mode and TEM were used for this study. Near the endurance limit, the dislocation cells smaller than 2£gm develop preferably along grain boundaries and triple junctions among the grains. Within grain interiors, it is hardly observed these small dislocation cells and cyclic hardening even at £`max =0.2%. When strain amplitudes were controlled at a range from £`max = 0.25% to 0.6%, a secondary cyclic hardening occurs prior to fatigue failure and less than 2um dislocation cells rapidly developed thoroughly. The secondary hardening rates were found to be directly proportional to the strain amplitudes. For high-low strain fatigue tests, while the maximum strain was decreased from 1.2% to 0.2% or 0.15%, dislocation cells were collapsed first and re-grouped into loop-patch structures due to the gliding behavior of dislocations changing from multiple-slips to single-slip. However, once the strain range is further reduced to 0.1%, dislocation cells would persist, showing no signs of collapse. Moreover, the reversal development of dislocation structures is independent of strain ratio. Furthermore newly developed loop patches are usually confined within dislocation domains with very condensed dislocation cell walls with high boundary misorientation.

Typescript (photocopy) / 132 leaves, [2] leaves of plates : ill. ; 30 cm. / Title page, contents and abstract only. The complete thesis in print form is available from the University Library. / Thesis (Ph.D.)--University of Adelaide, Dept. of Microbiology and Immunology, 1982