Showing papers on "Geometric data analysis published in 2005"

Face transfer with multilinear models

Abstract: Face Transfer is a method for mapping videorecorded performances of one individual to facial animations of another It extracts visemes (speech-related mouth articulations), expressions, and three-dimensional (3D) pose from monocular video or film footage These parameters are then used to generate and drive a detailed 3D textured face mesh for a target identity, which can be seamlessly rendered back into target footage The underlying face model automatically adjusts for how the target performs facial expressions and visemes The performance data can be easily edited to change the visemes, expressions, pose, or even the identity of the target---the attributes are separably controllable This supports a wide variety of video rewrite and puppetry applicationsFace Transfer is based on a multilinear model of 3D face meshes that separably parameterizes the space of geometric variations due to different attributes (eg, identity, expression, and viseme) Separability means that each of these attributes can be independently varied A multilinear model can be estimated from a Cartesian product of examples (identities × expressions × visemes) with techniques from statistical analysis, but only after careful preprocessing of the geometric data set to secure one-to-one correspondence, to minimize cross-coupling artifacts, and to fill in any missing examples Face Transfer offers new solutions to these problems and links the estimated model with a face-tracking algorithm to extract pose, expression, and viseme parameters

3d-modeling of urban structures

Abstract: Three dimensional building models have become important during the past years for various applications like urban planning, enhanced navigation or visualization of touristic or historic objects. 3D-models can increase the understanding and explanation of complex urban scenarios and support decision processes. A 3D-model of the urban environment gives the possibility for simulation and rehearsal, to "fly through" the local urban structures with multiple perspective viewing, and to visualize the scene out from different viewpoints. The building models are typically acquired by (semi-) automatic processing of Laser scanner elevation data or aerial imagery. We are presenting an automatic generation method of polyhedral 3D-models from Laser height data in our paper. The methods of deriving a DTM and a DSM from the data as well as the estimation of a ground map for the built-up area as alternative of a cadastral map are especially investigated. An approach for the classification of vegetation areas is presented. Although for some applications geometric data alone is sufficient, for visualization purposes a more realistic representation with textured surfaces is necessary. The associated textures from buildings are extracted either from airborne imagery or, especially for facades, from images taken by ground based cameras. We have investigated the selection of optimal texturing images from the acquired data including occlusions and multiple representations. Results are presented.

Automatic Building of Structured Geological Models

Abstract: The present article proposes a method to significantly improve the construction and updating of 3D geological models used for oil and gas exploration We present a prototype of a “geological pilot” which enables monitoring the automatic building of a 3D model topologically and geologically consistent, on which geological links between objects can easily be visualized This model can automatically be revised in case of changes in the geometric data or in the interpretation

Computer-integrated needle therapy systems: implementation and analysis

Abstract: As a class of problems, needle-based therapy applications are conducive targets for Computer-Integrated Surgery (CIS) Systems. While these applications have fairly consistent, simple guidance and manipulation requirements, the imaging, tracking, manipulation and data-processing technologies used to solve these problems can vary widely. This diversity proves to be a challenge. Improving pace and consistency in the development of prototype and full CIS systems would be of significant benefit to the broad community of scientists, engineers and doctors who develop these systems and use them to improve patients' lives. We address these issues by developing a modular framework for the design, implementation and analysis of CIS needle-therapy systems. In this dissertation we first present the underlying theory for this framework. This theoretical foundation provides the ability to both fully specify the geometric reasoning required for these systems and to automate the process of analyzing their resulting geometric accuracy. Building on this foundation, we present a system implementation tool which uses this theory to automate the process of building functional systems capable of collecting, processing, outputting and analyzing the geometric data necessary to implement and predict accuracies for a variety of needle-therapy approaches. We validate this work through the construction and analysis of several functional example systems, in both simulation and actual operation.

An ontology of virtual humans: incorporating semantics into human shapes

Abstract: Most of the efforts concerning graphical representations of humans (Virtual Humans) have been focused on synthesizing geometry for static or animated shapes. The next step is to consider a human body not only as a 3D shape, but as an active semantic entity with features, functionalities, interaction skills, etc. The ontology for Virtual Humans we are defining will provide the ”semantic layer” required to reconstruct, stock, retrieve and reuse content and knowledge related to Virtual Humans. The connection between the semantic and graphical data is achieved thanks to an intermediate layer based on anatomical features extracted form morphological shape analysis. The resulting shape descriptors can be used to derive higher-level descriptors from the raw geometric data. High-level descriptors can be used to control human models.

A study on modeling and analysis of kinematic motion deviations of machine tools (1st report, modeling and analysis of geometric deviations of constituting units)

Abstract: The objective of the present research is to develop a systematic method for planning and analysis of kinematic motion deviations of machine tools based on geometric tolerances of three-dimensional components. This paper deals with the modeling and analysis of the geometric deviations of the toleranced components. The tolerance zones of the geometric features of the components are investigated and classified based on the definitions of the geometric tolerances. The parameters are defined to describe the deviations of the geometric features within the tolerance zones. An analytical method is proposed to estimate the statistic deviations of the positions and the orientations of the geometric features, based on the deviation parameters and the relationships between the datum features and the toleranced features.

A formal description of preparation processes for geometric models to improve downstream activities integration in the design process

Abstract: In order to better integrate downstream applications - like simulation or virtual reality based applications - within design cycles, geometric data extracted from the digital mock-up need to be prepared and adapted to the needs and the constraints of these applications. In this paper, we propose to formalize the preparation process from the digital mock-up to an adapted database (AD). These processes allow not only transforming geometric data, but also creating a link between the initial data and the AD. The current computer implementation shows how this processes can be described and saved, in order to automate and capitalize processes, reduce preparation time, and finally enhance downstream applications integration within design.

General purpose cad device, and recording medium stored with program used for the device

Abstract: PROBLEM TO BE SOLVED: To provide a general purpose CAD device capable of supporting structure design and structure analysis of a building structure, estimation of the building structure, etc., based upon information prepared by the general purpose CAD device, and to provide a storage medium stored with a program used for the general purpose CAD device. SOLUTION: The general purpose CAD device is disclosed which has a geometric data storage area stored with geometric data generated by a drawing generating means for generating framework drawings of the building structure and their design external walls, etc., has a file area stored in prescribed record format with data regarding nodes such as the house or penthouse number, layer numbers, a reference layer, boundary condition number, node code number, etc., of the building structure, data regarding cross-sectional shape of members, data regarding load shapes, etc., based upon the geometric data, and has a structure planning reference file 8, a member shape reference file 9, and a load shape reference file 10 including the geometric data, and the storage medium having the program supporting the general purpose CAD device is also disclosed. COPYRIGHT: (C)2005,JPO&NCIPI

Flexible Terrain Representation using Runtime Surface Reconstruction

Abstract: Terrain has a great potential as a reference for visual navigation, which can be utilized to access and manage information. From this standpoint the geometry of the terrain is a unique defining surface for all geographic applications as well as for any geo-related information. However, a data representation of terrain in three dimensions provides numerous challenges for visualization as well as for analytical purposes. S olutions that satisfy visualization criterions often appear to be less optimal for maintenance or scalability required by analytical applications and vice versa. T his work proposes a geometric data representation of the terrain that respects both visualization and analytical applications. The solution can be used for an entire planet, which allows avoiding needs for performing conversions between cartographic projections and transformations between geodetic datums. The data representation provides good local geometric flexibility like TIN but also supports multiple levels of detail. Amounts of data can grow large gradually---it is possible to alter local areas only while leaving distant parts of the terrain unchanged, which is convenient for maintenance. Introduced approach is based on runtime construction of triangulated irregular network using Delaunay triangulation. The mass points used for the surface reconstruction are structured in order to support multiple levels-of-detail. The proposed representation allows managing terrain data in numerous detached repositories, which can be used for distributed solution.

Level of detail continuum for huge geometric data

Abstract: In this paper we propose a unified solution for the creation of levels of detail on very large input data. We build a hierarchical signed distance function in an octree around the data and use this hierarchy to generate a continuum of levels of detail. Our distance function construction, based on the Gradient Vector Flow and the Poisson equation, builds on multigrid resolution algorithms. Using an appropriate interpolation scheme within the octree we obtain a continuous hierarchical distance function, which allows us to define a continuum of levels of detail for huge geometries. During this process, holes and undersampling issues in the input data are automatically corrected. We present three applications of our hierarchy: a novel hierarchical deformable model scheme that can automatically reconstruct closed Eulerian meshes of up to a million faces in a few minutes, an alternate distance-driven contouring approach, and raytracing of huge data models.

Method for producing automotive parts

Abstract: The invention relates to a method for producing a bodywork component of a motor vehicle that has first been shaped and is then thermally surface-treated, said method comprising iterative simulated steps with variable geometric data of tools that are used in the shaping process. According to the invention, a continuous check verifies whether the expected geometric data of the component lies within the permitted tolerance range of the nominal geometric data of the component. If this is not the case, the geometric data and the generation of corrected tool geometric data is modified. This takes place in two stages, the first stage for the shaping steps and the subsequent stage for the process steps of the thermal surface treatment.

Voxel-based Solid Models: Representation, Display and Geometric Analysis

Abstract: i List of tables v List of figures vii

Forensic Retrieval of Striations on Fired Bullets by using 3D Geometric Data.

Abstract: Currently, optical devices, such as microscopes and CCD cameras, are mainly utilized for identification of bullets and tool marks in the field of forensic science. While these optical methods are easily manageable and efective, they are under great influence of illumination condition. Besides these appearance-based approaches, we can utilize 3D geometric data of tool marks that are free from lighting condition. Nevertheless, a perfect correspondence of two striation patterns is rarely encountered, even if the two bullets have been fired from the same firearm. Therefore, more robust methods are required. In this study, we propose a two-stage comparison method focused on 3D geometric. At first, we have aligned global shapes and evaluated a global shape similarity. Then small elevations are compared by neural networks; that is local matching. In this stage, rendered images under a unified lighting condition are utilized. By using 2-stage comparison, we have developed a robust method that searches for similar striation patterns.

Developmnet strategy of the road safety survey and analysis vehicle : rossav

Abstract: The objective of this research is to develop a strategy to build a mobile system which can be useful for the road safety analysis. Most commonly used data for the road safety analysis are road geometric information. In order to collect and analyze road geometric data, various sensors were tested and installed to collect position data, attitude data, and image data. The development strategies for the RoSSAV are suggested for the various sensors, signal synchronization devices, and the applications of sensor data. Based on the research performed, first, road data acquisition schemes were made; second, required sensors were defined; third, a proto-type analysis vehicle was developed; fourth, a signal synchronization device for the multiple-sensors were designed and developed; fifth, a three dimensional image analysis software was developed.

An Edge-Based Data Structure for Navier-Stokes Equations Resolution

Abstract: This work presents an edge-based data structure, that allows to compute solutions of Euler and Navier-Stokes equations, independently of the selected mesh type used to discretize the computational domain: The code developed, ’BERTA’, is grid-transparent. Based in this structure a complete solution technique has been formulated, which allows to handle structured grids, block structured grids, and unstructured grids of tethrahedral or mixed elements without any modification. The contribution of the work lie in the possibility of unbinding the information needed to solve the equetions from the geometric data and original dimension of the mesh. Allocating original geometric data to the mesh edge, subsequent references to the original grid are not needed. An efficient data structure has been constructed, minimizing memory overhead and amount of gather/scatter, in comparation to others structures (element-based or face-based). Furthemore, the gains in computational efficiency afforded by the use of hybrid meshes over fully tetrahedral meshes are demostrated. A finite-volumen scheme has been used for the spatial discretization of the equations, and multi-stage time-stepping scheme for the time discretization. Viability of this data structure to solve termofluiddynamics problems wiht large generality has been demonstrated.

Efficient Non-intersection Queries on Aggregated Geometric Data

Abstract: Let S be a set of geometric objects that are aggregated into disjoint groups. The problem considered is that of preprocessing S so that for any query object, q, the distinct groups such that no objects from those groups are intersected by q can be reported efficiently. The goal is to devise solutions where the query time is sensitive to the output size, i.e., the number of groups reported. Unfortunately, the obvious approaches of (i) solving the corresponding intersection problem for aggregated data and reporting the complement, or (ii) querying with the complement of q are either expensive or incorrect. Efficient, output-sensitive solutions are given to several non-intersection searching problems on aggregated data, using methods such as geometric duality, sparsification, persistence, filtering search, and pruning.

An integrated approach to design and analyze unprotected type flange coupling

Abstract: Most of mechanical CAD software provides mechanical drafting and design, which involves a variety of different types of drawings. All these drawings require coordinate–geometric data and descriptive data. The objective of this work is to establish software programs for designing, visualizing and analyzing of unprotected type flange coupling as a mechanical element of wide range of applications. The approach includes: (a) Data base of some carbon and alloy steel (used in manufacturing of flanges) with some of their corresponding mechanical properties, (b) Software program for the generation of geometric data of the flange coupling components (flange, hub, bolts and key set), (c) Data exchange file (DXF) for converting the calculated geometric data to a well known AutoCAD system for the purpose of visualization , (d) Model analysis by finite element method using MATLAB programming system to show the robust and weakness regions, and (e) A graphical user interface (GUI) to versatile and elegance the process.

Construction of a 3-d finite element model of the human forearm complex using a series of 2-d color images

Abstract: A simple yet efficient paradigm for geometric mesh generation using the Visible Human Project Male dataset for further finite element analysis was presented. The minimum distance classifier was used for the discriminant function between the class centers classified by the fuzzy c-means clustering method in the RGB space. Furthermore, based on two major geometric assumptions on the boundary curves, star-shaped polygon and geometric conformity, a points-on-line search technique was devised for efficient computation of the boundary points of the contours for each anatomical component of the human forearm complex. The computed boundary points in each slice were fitted to a closed spline curve and resampled and then refitted for correct alignment with the consecutive boundary curves in order to improve geometric fidelity. By using the refitted contours, a 3-D geometric model of the human radius, ulna, and surrounding soft tissue was generated in a commercial computer–aided design system and exported to a commercial finite element analysis package for meshing with its built-in automatic mesh generator. The proposed method can be applied to geometric mesh generation of other long bones, which allows easy handling, storage, and exchange of the model.

Non-conflicting method and system for three-dimensional geometric data

Abstract: PROBLEM TO BE SOLVED: To provide a method of detecting and eliminating geometric conflict (self-interference and the like) in three-dimensional geometric data, in particular, a geometric non-conflict processing method of the three-dimensional geometric data free from failure in operation even when the shape operation is repeatedly executed in the shape operation of three-dimensional geometric data. SOLUTION: This method of detecting and eliminating geometric conflict in a processed three-dimensional geometric data in a three-dimensional geometric data processing system SS performing the operation processing to the three-dimensional geometric data, and storing the geometric data created and processed in the operation processing process, comprises a step for constituting and processing the boundary face data as a surface data of a three-dimensional geometric figure by triangle face data or zero triangle data where three apexes of triangle are located on the same straight line, a coordinate value data step for arbitrarily truncating a numerical value of the three-dimensional geometric data in accordance with necessary accuracy, and a geometric conflict detecting and eliminating step for detecting and eliminating the geometric conflict generated in the numerical value data truncating step. COPYRIGHT: (C)2005,JPO&NCIPI

An efficient compression method for triangular meshes used in engineering

Abstract: This paper presents a new approach for the lossless compression of engineering data, represented by triangles The method works in two steps - a topology compression followed by an entropy compression The topology compression is based on the five states; one is automatically recognized by the compressor/decompressor and the remaining is coded by four commands This approach for topology compression has been compared with other methods and turns out to be highly competitive Geometric data and application-specific engineering data are also prepared for compression during the topology compression Namely, the compressed topological data contributes just a few percent to the normal size of the geometric and application-specific data General purpose compression methods are usually used for these We compared our method with the popular PkZip and we achieved considerably better results