VASC Seminar Announcement
        =========================

Speaker:   Owen Carmichael
Affiliation:  Carnegie Mellon University, Robotics Institute
E-mail:  otc@andrew.cmu.edu
Date: Monday, September 16
Time: 3:30pm
Place: NSH 1507

Title:   Detecting Wiry, Holey Objects In Cluttered Scenes

Abstract:

This talk will address the problem of detecting objects with wiry
components and holes in cluttered environments; an example application is
the recognition of chairs, lamps, and tables in an image of an office.  We
frame the problem in terms of edge cues: specifically, we determine
whether individual edge pixels in the image belong to the target object or
to clutter, based on the configuration of edges in their vicinity.  A
classifier solves this problem by computing sparse, localized edge
features at image locations determined at training time. In order to save
computation and solve the aperture problem, we apply a cascade of these
classifiers to the image, each of which computes edge features over larger
image regions than its predecessors. Experiments apply our approach to the
recognition of a chair and a cart in cluttered scenes under arbitrary
out-of-image-plane rotation.