By Topic

3-D Digital Imaging and Modeling, 2003. 3DIM 2003. Proceedings. Fourth International Conference on

Date 6-10 Oct. 2003

Filter Results

Displaying Results 1 - 25 of 63
  • Surface curvature estimation from the signed distance field

    Page(s): 361 - 368
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (684 KB) |  | HTML iconHTML  

    We propose a method of computing surface curvature properties from the signed distance field (SDF) samples in the 3D space. The SDF representation contains information of the surface normal at the closest point on the surface from the sampling point. The variance of these information from different sampling points within the neighborhood reflects the curvature information. Because this sampling is done in the 3D space, we do not directly referees to the parametric surface coordinates or polygon structures. The computation is stable because it requires only linear algebraic operations. It is possible to extract multiple scale curvatures by changing sampling interval. The proposed method was applied on real data, and result of multiscale curvature extraction is presented. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Geometrically stable sampling for the ICP algorithm

    Page(s): 260 - 267
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1820 KB) |  | HTML iconHTML  

    The iterative closest point (ICP) algorithm is a widely used method for aligning three-dimensional point sets. The quality of alignment obtained by this algorithm depends heavily on choosing good pairs of corresponding points in the two datasets. If too many points are chosen from featureless regions of the data, the algorithm converges slowly, finds the wrong pose, or even diverges, especially in the presence of noise or miscalibration in the input data. We describe a method for detecting uncertainty in pose, and we propose a point selection strategy for ICP that minimizes this uncertainty by choosing samples that constrain potentially unstable transformations. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Active balloon model based on 3D skeleton extraction by competitive learning

    Page(s): 87 - 94
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (510 KB) |  | HTML iconHTML  

    We focus on the polygonal representation of a 3D object model which is composed of a lot of points on the object surface. In the 3D animation, it is sometimes necessary to use multiresolution representation of a model to cope with various situations, and it is preferable to represent the models with different resolutions in a certain unified data structure. In order to meet these requirements, we present a new method which generates the approximated model of an original one by using multiple deformable models, called active balloon models (ABMs). We extract the three-dimensional skeleton of the object. The obtained skeleton comprises nodes and edges, and represents the structure of the object. Based on the skeleton, the approximated model is generated by deforming the ABMs. Some experimental works are made to verify the capability of our method. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Registration and fusion of intensity and range data for 3D modelling of real world scenes

    Page(s): 418 - 425
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (2067 KB) |  | HTML iconHTML  

    A novel technique combining intensity and range data is presented. Passive (intensity based) and active (range based) techniques used for 3D reconstruction have their limitations and separately, none of these techniques can solve all the problems inherent to the modelling of real environments. Our technique aims to demonstrate how both intensity and range data can be registered and combined into a long-range 3D system. The procedure needs an initial estimation for internal and external camera parameters for two or more intensity images. The technique uses passive triangulation of the intensity data to refine the initial camera calibrations and ensure a good registration of range and video data sets. Once a reliable calibration is achieved, corresponding points from the intensity images are triangulated and introduced in the original range data. With our technique, it is possible to complete the models in areas where data is missing or to increase the resolution in areas of high interest and 3D contents. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The ModelCamera: a hand-held device for interactive modeling

    Page(s): 285 - 292
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1611 KB) |  | HTML iconHTML  

    An important goal of automated modeling is to provide computer graphics applications with high quality models of complex real-world scenes. Prior systems have one or more of the following disadvantages: slow modeling pipeline, applicability restricted to small scenes, no direct color acquisition, and high cost. We describe a hand-held scene modeling device that operates at five frames per second and that costs $2,000. The device consists of a digital video camera with 16 laser pointers attached to it. As the operator scans the scene, the pointers cast blobs that are detected and triangulated to provide sparse, evenly spaced depth samples. The frames are registered and merged into an evolving model, which is rendered continually to provide immediate operator feedback. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A discrete Reeb graph approach for the segmentation of human body scans

    Page(s): 378 - 385
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1193 KB) |  | HTML iconHTML  

    Segmentation of 3D human body (HB) scan is a very challenging problem in applications exploiting human scan data. To tackle this problem, we propose a topological approach based on discrete Reeb graph (DRG) which is an extension of the classical Reeb graph to unorganized cloud of 3D points. The essence of the approach is detecting critical nodes in the DRG thus permitting the extraction of branches that represent the body parts. Because the human body shape representation is built upon global topological features that are preserved so long as the whole structure of the human body does not change, our approach is quite robust against noise, holes, irregular sampling, moderate reference change and posture variation. Experimental results performed on real scan data demonstrate the validity of our method. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Segmentation and modeling of approximately rotationally symmetric objects in 3D ultrasound

    Page(s): 124 - 131
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1788 KB)  

    We describe a method for segmentation and modeling of approximately rotationally symmetric objects in 3D ultrasound. The proposed algorithm is based on 2D segmentation of several slices of an ultrasound volume dataset. It requires semiautomatic 2D segmentation of only two slices by an interactive process which is based on a discrete dynamic contour. A tiling algorithm leads to a 2D starting contour in every slice. A fully automatic segmentation process is then performed to all slices simultaneously. The result is a triangular mesh of the segmented object. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Human figure reconstruction and modeling from single image or monocular video sequence

    Page(s): 116 - 123
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1576 KB) |  | HTML iconHTML  

    We firstly review some approaches to recover the 3D shape of human figures and the related movements. Then we present an improved and reliable version of an easy approach to recover 3D model of humans using just one frame or a monocular video sequence. A simplification of the camera model based on the collinearity condition is required, due to the absence of stereo view. This leads to a perspective or orthographic projection and requires additional information or constraints to solve the problem. The human figure is reconstructed in a skeleton form and visualized with the portable VRML language. Moreover a surface fitting is performed using a predefined 3D model obtained from laser scanner measurements. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Solving architectural modelling problems using knowledge

    Page(s): 343 - 351
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (619 KB) |  | HTML iconHTML  

    We summarize a series of recent research results made at Edinburgh University based on applying domain knowledge of standard shapes and relationships to solve or improve architectural reconstruction problems. The problems considered are how to enforce known relationships when data fitting, how to extract features even in very noisy data, how to get better shape parameter estimates and how to infer data about unseen features. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A hierarchical method for aligning warped meshes

    Page(s): 434 - 441
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1211 KB) |  | HTML iconHTML  

    Current alignment algorithms for registering range data captured from a 3D scanner assume that the range data depicts identical geometry taken from different views. However, in the presence of scanner calibration errors, the data will be slightly warped. These warps often cause current alignment algorithms to converge slowly, find the wrong alignment, or even diverge. We present a method for aligning warped range data represented by polygon meshes. Our strategy can be characterized as a coarse-to-fine hierarchical approach, where we assume that since the warp is global, we can compensate for it by treating each mesh as a collection of smaller piecewise rigid sections, which can translate and rotate with respect to each other. We split the meshes subject to several constraints, in order to ensure that the resulting sections converge reliably. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Efficient surface reconstruction from range curves

    Page(s): 54 - 61
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (519 KB) |  | HTML iconHTML  

    We propose an approach for surface reconstruction of free-form rigid objects from an arbitrary set of intersecting range curves. A strategy for updating the reconstructed surface during data acquisition is described as well. Geometric and color information is accumulated in a volumetric structure in which a vector field is built and updated. Moreover, the information that is needed for efficient curve registration is directly available in this vector field. This leads to a unified modeling approach combining surface reconstruction and curve registration. The algorithm implementing the approach is of linear complexity with respect to the number of input curves and makes it suitable for interactive modeling. A compression scheme based on a multiresolution decomposition of vector fields is introduced as well. Simulated data from a set of curvilinear patterns as well as data acquired with a hand-held range sensor are used to validate the approach. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Cramer-Rao bounds for nonparametric surface reconstruction from range data

    Page(s): 70 - 77
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (592 KB) |  | HTML iconHTML  

    The Cramer-Rao error bound provides a fundamental limit on the expected performance of a statistical estimator. The error bound depends on the general properties of the system, but not on the specific properties of the estimator or the solution. The Cramer-Rao error bound has been applied to scalar- and vector-valued estimators and recently to parametric shape estimators. However, nonparametric, low-level surface representations are an important tool in 3D reconstruction, and are particularly useful for representing complex scenes with arbitrary shapes and topologies. We present a generalization of the Cramer-Rao error bound to nonparametric shape estimators. Specifically, we derive the error bound for the full 3D reconstruction of scenes from multiple range images. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Taking consensus of signed distance field for complementing unobservable surface

    Page(s): 410 - 417
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1853 KB) |  | HTML iconHTML  

    When we use range finders to observe the shape of an object, many occluded areas may occur. These become holes and gaps in the model and make it undesirable to utilize the model for various applications. We propose a novel method to fill holes and gaps and complement such an incomplete model. We use a signed distance field (SDF) as an intermediate representation, which stores Euclidean signed distances from a voxel to the nearest point of the mesh model. Since the signs of a SDF become unstable around holes or gaps, we take a consensus of the signed distances of neighbor voxels by estimating the consistency of the SDF. Once we make the SDF consistent, we can efficiently fill holes and gaps. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Efficient reconstruction of indoor scenes with color

    Page(s): 402 - 409
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1777 KB) |  | HTML iconHTML  

    We present an efficient and general approach to computing and integrating 3D distance fields directly from multiple range images. We compute normal and confidence values directly from a 2D range image. We then approximate 3D Euclidean distance by correcting the line-of-sight distance. To integrate multiple scans, we efficiently transform voxels of the target distance field to each scan's local coordinate system, then update voxels with computed distance and confidence values. Finally, we extract an iso-surface from the weighted distance field using the marching cubes algorithm. We extend the same idea to the assignment of weighted colors or texture coordinates to the reconstructed model. Experiments show that our approach is fast, has reasonable storage requirements, and can produce high-quality surfaces from multiple range scans. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Silhouette and stereo fusion for 3D object modeling

    Page(s): 46 - 53
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1140 KB) |  | HTML iconHTML  

    We present a new approach to high quality 3D object reconstruction. Starting from a calibrated sequence of color images, the algorithm is able to reconstruct both the 3D geometry and the texture. The core of the method is based on a deformable model, which defines the framework where texture and silhouette information can be fused. This is achieved by defining two external forces based on the images: a texture driven force and a silhouette driven force. The texture force is computed in two steps: a multistereo correlation voting approach and a gradient vector flow diffusion. Due to the high resolution of the voting approach, a multigrid version of the gradient vector flow has been developed. Concerning the silhouette force, a new formulation of the silhouette constraint is derived. It provides a robust way to integrate the silhouettes in the evolution algorithm. As a consequence, we are able to recover the apparent contours of the model at the end of the iteration process. Finally, a texture map is computed from the original images for the reconstructed 3D model. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • 3D animation of cerebral activity using both spatial and temporal fMRI information

    Page(s): 103 - 109
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (559 KB) |  | HTML iconHTML  

    Functional MRI (fMRI) has become a major tool in understanding brain function. Functional maps are superimposed on higher definition anatomical images. However these maps are only visualized in 2D or 3D. The temporal information, provided by the fMRI acquisition, is usually ignored. The main purpose is to present a software designed to provide animation of activation maps. This approach helps researchers to follow temporal dynamics of the neuronal network involved in brain activation at different time scales, allowing them to identify the priority of the activation of cerebral structures. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Scanning and processing 3D objects for Web display

    Page(s): 310 - 317
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (4586 KB)  

    We present a case study of scanning 3D objects for the purposes of education and public information. We begin by describing the original design of a 3D scanning system now in use in Cairo's Egyptian Museum. The system captures both the geometry and surface color and detail of museum artifacts. We report on the experience using the system in the museum setting, and how practical problems with the system were addressed. We present samples of how the processed 3D data will be used on a Web site designed to communicate Egyptian culture. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Combining texture and shape for automatic crude patch registration

    Page(s): 179 - 186
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1137 KB) |  | HTML iconHTML  

    Building complete three-dimensional models often requires the registration and integration of multiple partial reconstructions. Registration is usually performed using a coarse-to-fine approach. We present a novel method for the crude registration of many partial three-dimensional reconstructions (patches). Our method combines the 3D measurements with the texture. Registration with methods that use only geometry will fail if the shape is too simple or symmetric. In such cases one has to use texture to resolve the ambiguities. But even then, knowledge of the shape greatly simplifies the task. The strategy we use is to intersect the surface with small spheres centered at feature points. The intersection line between such a sampling sphere and the surface defines an invariant region in the texture image. The surface texture inside these regions is used for matching regions between different patches. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Virtual reconstruction of broken and unbroken pottery

    Page(s): 318 - 325
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1310 KB) |  | HTML iconHTML  

    Motivated by the requirements of the present archaeology, we are developing an automated system for archaeological classification and reconstruction of ceramics. Due to the nature of ceramics, most of the excavated vessels are in the form of fragments called sherds. Only a few of the finds are complete, however these finds are the most important and interesting ones. Therefore we are developing a system that handles both complete and broken vessels using two different reconstruction strategies: A shape from silhouette based method for complete vessels and a profile based method for fragments. The profile is the cross-section of the fragment in the direction of the rotational axis of symmetry and can be represented by a closed curve in the plane. For complete vessels the 3D reconstruction is based on a sequence of images of the object taken from different viewpoints. Then the output of both algorithms is used to construct the 3D model of the vessel for classification and display. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Optimized position sensors for flying-spot active triangulation systems

    Page(s): 28 - 36
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (530 KB) |  | HTML iconHTML  

    A description of the integrated sensors developed for flying-spot active triangulation is given. All the sensors have been fabricated using standard CMOS technology that allows the monolithic integration of photo-sensors, together with readout circuits, and digital signal processors. Position sensors are classified into two classes that allow a better understanding of the pros and cons of each one. A description of the proposed position sensor that is optimized for accurate and fast 3D acquisition is given alongside some experimental results. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Recursive model optimization using ICP and free moving 3D data acquisition

    Page(s): 251 - 258
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (817 KB) |  | HTML iconHTML  

    We describe a recursive multiresolution algorithm that reconstructs high-resolution and high-accuracy 3D images from low-resolution sparse range images or profiles. The method starts by creating a rough, partial, and potentially distorted estimate of the model of the object from an initial subset of sparse range data; then, using ICP algorithms, it recursively improves and refines the model by adding new range information. In parallel, real-time tracking of the object is performed in order to allow the laser scan to be automatically centered on the object. The end result is the creation of a high-resolution and accurate 3D model of a free-floating object, and real-time tracking of its position. Examples of the method are presented when the object and the 3D camera are moving freely with respect to each other. The system provides high accuracy hand-held laser scanning that does not require complex and costly mechanical scanning apparatus or external positioning devices. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Automatic 3D modelling of palatal plaster casts

    Page(s): 132 - 138
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (999 KB) |  | HTML iconHTML  

    We introduce a procedure for automatic 3D modelling and discusses its performance in the context of a medical application, namely the 3D modelling of palate's casts. The general interest of the proposed method lays in the fact that a widespread use of 3D modelling in nonengineering applications requires automatic procedures. The specific interest is in the intriguing mixture of open issues concerning 3D modelling, geometrical metrology and medicine represented by the modelling of free-form surfaces of anatomical interest. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Adaptive enhancement of 3D scenes using hierarchical registration of texture-mapped 3D models

    Page(s): 203 - 210
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (871 KB) |  | HTML iconHTML  

    Adaptive fusion of new information in a 3D urban scene is an important goal to achieve in computer vision, graphics, and visualization. We acquire new image pairs of a scene from closer distances and extract 3D models of successively higher resolutions. We present a new hierarchical approach to register these texture-mapped 3D models with a coarse 3D texture mapped model of an urban scene. First, we use the standard reconstruction algorithm to construct 3D models after establishing 1-1 correspondence between the feature points of two images at same resolution. Next, a subset of these feature points is used to register the higher resolution image with the lower resolution image using a scale-sensitive algorithm. Finally we register and consistently merge the 3D models at different resolutions. We present the results of our hierarchical algorithm for adaptive enhancement of a mural inside the UCSC Campus by registering data that differ in scale by a ratio of 1:15. Results indicate that the proposed hierarchical registration technique effectively utilizes the intermediate models to enable the smooth registration of the high resolution models on the coarser models. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A multi-resolution ICP with heuristic closest point search for fast and robust 3D registration of range images

    Page(s): 427 - 433
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (393 KB) |  | HTML iconHTML  

    The iterative closest point (ICP) algorithm is widely used for the registration of 3D geometric data. One of the main drawbacks of the algorithm is its quadratic time complexity O(N2) with the number of points N. Consequently, several methods have been proposed to accelerate the process. We present a new solution for the speeding up of the ICP algorithm and special care is taken to avoid any tradeoff with the quality of the registration. The proposed solution combines a coarse to fine multiresolution approach with the neighbor search algorithm. The multiresolution approach permits to successively improve the registration using finer levels of representation and the neighbor search algorithm speeds up the closest point search by using a heuristic approach. Both multiresolution scheme and neighbor search algorithm main features are presented. Confirming the success of the proposed solution, typical results show that this combination permits to create a very fast ICP algorithm, with a closest point search complexity of O(N), while preserving the matching quality. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Deformable model with adaptive mesh and automated topology changes

    Page(s): 12 - 19
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1650 KB) |  | HTML iconHTML  

    Due to their general and robust formulation deformable models offer a very appealing approach to 3D image segmentation. However there is a trade-off between model genericity, model accuracy and computational efficiency. In general, fully generic models require a uniform sampling of either the space or their mesh. The segmentation accuracy is thus a global parameter. Recovering small image features results in heavy computational costs whereas generally only restricted parts of images require a high segmentation accuracy. We present a highly deformable model that both handles fully automated topology changes and adapts its resolution locally according to the geometry of image features. The main idea is to replace the Euclidean metric with a Riemannian metric that expands interesting parts of the image. Then, a regular sampling is maintained with this new metric. This allows to automatically handle topology changes while increasing the model resolution locally according to the geometry of image components. By this way high quality segmentation is achieved with reduced computational costs. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.