Lefschetz Center for Dynamical Systems Seminar
**Special Brown Applied Mathematics Pattern Theory and Vision Seminar
Abstract: Gaile Gordon from Interval Research Corp. (Palo Alto) will give an informal talk about her work on face recognition using both depth data and curvature descriptors (thesis work with David Mumford), 2D intensity images, and video sequences (DARPA/ARL FERET Program). If time permits the talk will also touch on current work at Interval involving modification of spoken content in video sequences.
Brown University Center for Statistical Sciences Seminar
Abstract: Kaplan-Meier curves provide descriptors of survival information for different subgroups within a data set. Investigators frequently use these figures for descriptive comparison of the effect of a particular measure upon survival. When subjects enter a study at different ages and age is associated with survival, it is often desirable to adjust survival information for age differences in the subgroups. This talk will discuss several methods for age-adjustment of survival curves, including direct age-adjustment and proportional hazard modeling with age as a covariate. Examples will be presented from simulated data and from information collected in the Framingham Heart Study.
Brown Analysis Seminar
Special Lefschetz Center for Dynamical Systems Seminar
Abstract: It is shown, by constructing an appropriate example, that delay differential equations with negative feedback can possess asymptotically stable rapidly oscillating solutions. The example involves a piecewise constant nonlinearity which can be made a smooth function without destroying the dynamics. Asymptotic behavior in the delay equation with the piecewise constant nonlinearity admits an exact reduction to a 3-dimensionnal map which has an attracting fixed point corresponding to a periodic soluton of the differential equation.
Special Brown Applied Mathematics Pattern Theory and Vision Seminar
Abstract: Artificial Intelligence will sit in an iron cage as long as not being able to perceive and act in natural environments. This goal has proved very elusive. We have set out to break the cage open by concentrating on a key problem, object vision: the detection and interpretation of coherent objects in the environment. We achieve object detection and segmentation by integrating several cues, among them motion, color, stereo, boundary detection and form discrimination. We achieve invariant and robust object interpretation using elastic matching and visual data representation in terms of wavelets. I will illustrate our achievements with results from concrete projects, among them person recognition from video input and flexible robot object grasping under visual control. I will conclude by discussing the implementation of our mechanisms in the visual system and the biological relevance of our results.
PDE Seminar
LEMS and Electrical Science Seminar
Abstract: Three dimensional Magnetic Resonance (MR), and Spiral Computed Tomography (CT) techniques are now available on the imaging equipment found in most Radiology Departments in the U.S. These techniques have reduced acquisition time and improved image quality. Recently, surgeons have discovered that these patient data sets, when coupled with high speed computers and three dimensional digitizers, can provide valuable information in the operating room. An overview of the current clinical capability of these systems will be presented, along with a description of their components. The future of computer assisted surgery will be discussed.
<--- 1997 Index