Object recognition from sensory data involves, in part, determining the pose of a model with respect to a scene. A common method for finding an object's pose is the generalized Hough transform, which accumulates evidence for possible coordinate transformations in a parameter space whose axes are the quantized transformation parameters. Large clusters of similar transformations in that space are taken as evidence of a correct match. In this article, we provide a theoretical analysis of the behavior of such methods. We derive bounds on the set of transformations consistent with each pairing of data and model features, in the presence of noise and occlusion in the image. We also provide bounds on the likelihood of false peaks in the parameter space, as a function of noise, occlusion, and tessellation effects. We argue that blithely applying such methods to complex recognition tasks is a risky proposition, as the probability of false positives can be very high.

Hierarchical clustering is a common method used to determine clusters of similar data points in multidimensional spaces. O(n 2 ) algorithms are known for this problem [3, 4, 10, 18]. This paper reviews important results for sequential algorithms and describes previous work on parallel algorithms for hierarchical clustering. Parallel algorithms to perform hierarchical clustering using several distance metrics are then described. Optimal PRAM algorithms using n log n processors are given for the average link, complete link, centroid, median, and minimum variance metrics. Optimal butterfly and tree algorithms using n log n processors are given for the centroid, median, and minimum variance metrics. Optimal asymptotic speedups are achieved for the best practical algorithm to perform clustering using the single link metric on a n log n processor PRAM, butterfly, or tree. Keywords. Hierarchical clustering, pattern analysis, parallel algorithm, butterfly network, PRAM algorithm. 1 In...

We present an automatic method for docking organic ligands into protein Center for Information binding sites. The method can be used in the design process of specific Technology (GMD), Institute protein ligands. It combines an appropriate model of the physico-chemical for Algorithms and Scientific properties of the docked molecules with efficient methods for sampling the Computing (SCAI), Schloß conformational space of the ligand. If the ligand is flexible, it can adopt

: Some medical interventions require knowing the correspondence between an MRI/CT image and the actual position of the patient. Examples occur in neurosurgery and radiotherapy, but also in video surgery (laparoscopy). We present in this paper three new techniques for performing this task without artificial markers. To do this, we find the 3D2D projective transformation (composition of a rigid displacement and a perspective projection) which maps a 3D object onto a 2D image of this object. Depending on the object model (curve or surface), and on the 2D image acquisition system (X-Ray, video), the techniques are different but the framework is common: ffl We first find an estimate of the transformation using bitangent lines or bitangent planes. These are first order semi-differential invariants [GMPO92]. ffl Then, introducing the normal or tangent, we define a distance between the 3D object and the 2D image, and we minimize it using extensions of the Iterative Closest Point algorithm (...

Many recent object recognition systems use a small number of pairings of data and model features to compute the 3D transformation from a model coordinate frame into the sensor coordinate system. In the case of perfect image data, these systems seem to work well. With uncertain image data, however, the performance of such methods is less well understood. In this paper, we examine the effects of two-dimensional sensor uncertainty on the computation of three-dimensional model transformations. We use this analysis to bound the uncertainty in the transformation parameters, as well as the uncertainty associated with applying the transformation to map other model features into the image. We also examine the effects of the transformation uncertainty on the effectiveness of recognition methods.

This dissertation presents an integrated human shape modeling, detection, and body part localization vision system. It demonstrates that the system can (1) detect pedestrians in various shapes, sizes, postures, partial occlusion, and clothing from a moving vehicle using stereo cameras; (2) locate the joints of a person automatically and accurately without employing any markers around the joints.

Model-based pose estimation techniques that match image and model triangles rquire large numbers of matching operations in realworld applications. We show that by using approximations to perspective, 2-D lookup tables can be built for each of the triangles of the models. An approximation called "weak perspective" has been applied previously to this problem; we consider two other perspective approximations: paraperspective and orthoperspective. These approximations produce lower errors for off-center image features than weak perspective.

Industrial assembly involves sensing the pose (orientation and position) of a part. Efficient and reliable sensing strategies can be developed for an assembly task if the shape of the part is known in advance. In this article we investigate two problems of determining the pose of a polygonal part of known shape for the cases of a continuum and a finite number of possible poses respectively. The first problem, named sensing by inscription, involves determining the pose of a convex n-gon from a set of m supporting cones. An algorithm with running time O(nm) that almost always reduces to O(n+m log n) is presented to solve for all possible poses of the polygon. We prove that the number of possible poses cannot exceed 6n, given m ≥ 2 supporting cones with distinct vertices. Simulation experiments demonstrate that two supporting cones are sufficient to determine the real pose of the n-gon in most cases. Our results imply that sensing in practice can be carried out by obtaining viewing angles of a planar part at multiple exterior sites in the plane. On many occasions a parts feeder will have reduced the number of possible poses of a part to a small finite set. Our second problem, named sensing by point sampling,

David Lowe's influential and classic algorithm for tracking objects with known geometry is formulated with certain simplifying assumptions. A version implemented by Ishii et al. makes different simplifying assumptions. We formulate a full projective solution and apply the same algorithm (Newton's method). We report results of extensive testing of these three algorithms. We compute two image--space and six pose--space error metrics to quantify the effects of object pose, errors in initial solutions, and image noise levels. We consider several scenaria, from relatively unconstrained conditions to those that mirror real--world and real-- time constraints. The conclusion is that the full projective formulation makes the algorithm orders of magnitude more accurate and gives it super--exponential convergence properties with arguably better computation--time properties. This material is based on work supported by the Luso--American Foundation, Calouste Gulbenkian Foundation, JNICT, CAPES pro...

Tracking, or camera pose determination, is the main technical challenge in creating augmented realities. Constraining the degree to which the environment may be altered to support tracking heightens the challenge. This paper describes several years of work at the USC Computer Graphics and Immersive Technologies (CGIT) laboratory to develop self-contained, minimally intrusive tracking systems for use in both indoor and outdoor settings. These hybrid-technology tracking systems combine vision and inertial sensing with research in fiducial design, feature detection, motion estimation, recursive filters, and pragmatic engineering to satisfy realistic application requirements.