🚀 go-pugleaf

Date:    3 March 2000
Time:    2:00 p.m.
Place:   Wean Hall 4623
Type:    Thesis Oral
Who:     Henry Schneiderman
Topic:   3D Object Detection using a "Maximalist" Statistical
         Representation: Faces and Cars


Abstract:

In this thesis, we describe a statistical method for 3D object
detection.   We represent the statistics of both object appearance and
"non-object" appearance using a product of histograms.  Each histogram
represents the joint statistics of a subset of wavelet coefficients and
their position on the object.  Our approach is to use many such
histograms representing a wide variety of visual attributes.  Using this
method, we have developed the first algorithm that can reliably detect
faces that vary from frontal view to full profile view and the first
algorithm that can reliably detect cars over a wide range of viewpoints.
By using coarse-to-fine heuristics, the face detector executes at 90
seconds per image and the car detector at 450 seconds per image on
240x256 images using a Pentium II 450Mhz CPU.


Further Details: 

Committee Members:  Takeo Kanade, Chair
                    Tai-Sing Lee
                    Alex Pentland, MIT Media Lab
                    Dean Pomerleau, AssistWare Technology


A copy of the thesis document is available at: 

/afs/cs.cmu.edu/user/hws/thesis.ps