A novel scheme for the detection of object boundaries is presented. The technique is based on active contours evolving in time according to intrinsic geometric measures of the image. The evolving contours naturally split and merge, allowing the simultaneous detection of several objects and both interior and exterior boundaries. The proposed approach is based on the relation between active contours and the computation of geodesics or minimal distance curves. The minimal distance curve lays in a Riemannian space whose metric is defined by the image content. This geodesic approach for object segmentation allows to connect classical "snakes" based on energy minimization and geometric active contours based on the theory of curve evolution. Previous models of geometric active contours are improved, allowing stable boundary detection when their gradients suffer from large variations, including gaps. Formal results concerning existence, uniqueness, stability, and correctness of the evolution are presented as well. The scheme was implemented using an efficient algorithm for curve evolution. Experimental results of applying the scheme to real images including objects with holes and medical data imagery demonstrate its power. The results may be extended to 3D object segmentation as well.

This paper describes a technique for building compact models of the shape and appearance of flexible objects (such as organs) seen in 2-D images. The models are derived from the statistics of sets of labelled images of examples of the objects.

We present an approach to modeling with truly mutable yet completely controllable free-form surfaces of arbitrary topology. Surfaces may be pinned down at points and along curves, cut up and smoothly welded back together, and faired and reshaped in the large. This style of control is formulated as a constrained shape optimization, with minimization of squared principal curvatures yielding graceful shapes that are free of the parameterization worries accompanying many patch-based approaches. Triangulated point sets are used to approximate these smooth variational surfaces, bridging the gap between patch-based and particle-based representations. Automatic refinement, mesh smoothing, and re-triangulation maintain a good computational mesh as the surface shape evolves, and give sample points and surface features much of the freedom to slide around in the surface that oriented particles enjoy. The resulting surface triangulations are constructed and maintained in real time. 1 Introduction ...

Motivated by Lagrangian simulation of elastic deformation, we propose a new tetrahedral mesh generation algorithm that produces both high quality elements and a mesh that is well conditioned for subsequent large deformations. We use a signed distance function defined on a Cartesian grid in order to represent the object geometry. After tiling space with a uniform lattice based on crystallography, we use the signed distance function or other user defined criteria to guide a red green mesh subdivision algorithm that results in a candidate mesh with the appropriate level of detail. Then, we carefully select the final topology so that the connectivity is suitable for large deformation and the mesh approximates the desired shape. Finally, we compress the mesh to tightly fit the object boundary using either masses and springs, the finite element method or an optimization approach to relax the positions of the nodes. The resulting mesh is well suited for simulation since it is highly structured, has robust topological connectivity in the face of large deformations, and is readily refined if deemed necessary during subsequent simulation.

The automatic segmentation and labelling of anatomical structures in 3D medical imagesis a challenging task of practical importance. We describe amodel- based approach which allows robust and accurate interpretation using explicit anatomical knowledge. Our method is based on the extension to 3D of Point Distribution Models (PDMs) and associated image search algorithms. Acombination of global, GeneticAlgorithm (GA), and local, Active Shape Model (ASM), search is used. We have built a 3D PDM of the human brain describing a number of major structures. Using this model we have obtained automatic interpretations for 30 3D Magnetic Resonance head images from different individuals. The results have been evaluated quantitativelyand support our claim of robust and accurate interpretation.

The analysis and automatic interpretation of images containing moving non-rigid objects, such as walking people, has been the subject of considerable research in the field of computer vision and pattern recognition. In order to build fast and reliable systems some kind of prior model is generally required. A model enables the system to cope with situations where there is considerable background clutter or where information is missing from the image data. This may be due to imaging errors (e.g. bluring due to motion) or due to part of an object becoming hidden from view. Conventional approaches to the problem of tracking non-rigid objects require complex handcrafted models which are not easily adapted to different problems. A more recent approach uses training information to build models for image analysis. This thesis extends this approach by building flexible 2D models, automatically, from sequences of training images. Efficient methods are described for using the resulting models for rea...

We present a tetrahedral mesh generation algorithm designed for the Lagrangian simulation of deformable bodies. The algorithm’s input is a level set (i.e., a signed distance function on a Cartesian grid or octree). First a bounding box of the object is covered with a uniform lattice of subdivision-invariant tetrahedra. The level set is then used to guide a red green adaptive subdivision procedure that is based on both the local curvature and the proximity to the object boundary. The final topology is carefully chosen so that the connectivity is suitable for large deformation and the mesh approximates the desired shape. Finally, this candidate mesh is compressed to match the object boundary. To maintain element quality during this compression phase we relax the positions of the nodes using finite elements, masses and springs, or an optimization procedure. The resulting mesh is well suited for simulation since it is highly structured, has topology chosen specifically for large deformations, and is readily refined if required during subsequent simulation. We then use this algorithm to generate meshes for the simulation of skeletal muscle from level set representations of the anatomy. The geometric complexity of biological materials makes it very difficult to generate these models procedurally and as a result we obtain most if not all data from an actual human subject. Our current method involves using voxelized data from the Visible Male [1] to create level set representations of muscle and bone geometries. Given this representation, we use simple level set operations to rebuild and repair errors in the segmented data as well as to smooth aliasing inherent in the voxelized data.

We present Easy Mesh Cutting, an intuitive and easy-to-use mesh cutout tool. Users can cut meaningful components from meshes by simply drawing freehand sketches on the mesh. Our system provides instant visual feedback to obtain the cutting results based on an improved region growing algorithm using a feature sensitive metric. The cutting boundary can be automatically optimized or easily edited by users. Extensive experimentation shows that our approach produces good cutting results while requiring little skill or effort from the user and provides a good user experience. Based on the easy mesh cutting framework, we introduce two applications including sketch-based mesh editing and mesh merging for geometry processing. Categories and Subject Descriptors (according to ACM CCS): I.3.5 [Computer Graphics]: Geometric algorithms, languages, and systems 1.

Computer sensing of hand and limb motion is an important problem for applications in human-computer interaction, virtual reality, and athletic performance measurement. We describe a framework for local tracking of self-occluding motion, in which parts of the mechanism obstruct each others visibility to the camera. Our approach uses a kinematic model to predict occlusion and windowed templates to track partially occluded objects. We analyze our model of self-occlusion, discuss the implementation of our algorithm, and give experimental results for 3D hand tracking under significant amounts of self-occlusion. These results extend the DigitEyes system for articulated tracking described in [22, 21] to handle self-occluding motions. Contents 1 Introduction 1 2 A Framework for Tracking Self-Occluding Objects 2 2.1 Representing Self-Occlusion with Layered Templates : : : : : : : : : : : : : : 3 2.2 Local Visibility Orders for Templates : : : : : : : : : : : : : : : : : : : : : : 4 2.3 Wind...

In this article we present a new method for visual face tracking that is carried out in wavelet subspace. Firstly, a wavelet representation for the face template is created, which spans a low dimensional subspace of the image space. The wavelet representation of the face is a point in this wavelet subspace. The video sequence frames in which the face is tracked are orthogonally projected into this low-dimensional subspace. This can be done efficiently through a small number of local projections of the wavelet functions. All further computations are then performed in the low-dimensional subspace. The wavelet subspace inherits its invariance to rotation, scale and translation from the wavelets; shear invariance can also be achieved, which makes the subspace invariant to affine deformations.