BIOMETRICS

Fall 2012

 

 

 

 

 

Course Information

Biometrics:  CS 4737 and CS 6737

Classroom: 627 MUDD
Times: T TH  1:10 pm – 2:25 pm

Web page : http://www1.cs.columbia.edu/~belhumeur/courses/biometrics/2012/biometrics.html
 

 

Instructor 

Prof. Peter N. Belhumeur
http://www.cs.columbia.edu/~belhumeur/
Email: belhumeur@cs.columbia.edu
Office: 623 CEPSR
Phone: 212-939-7087

Office Hours: Mon 8:30-10:00am

 

TAs
Jiongxin Liu
Email: liujx09@cs.columbia.edu

Office: CEPSR 6LW4

Office Hours:  Mon 4:30-6:30pm

 

Thomas Berg
Email: tberg@cs.columbia.edu

Office: CEPSR 6LW4

Office Hours:  Wed 3:00-5:00pm

 

 

 

Overview

The earliest known use of biometrics dates back to the 7th century during China's Tang Dynasty; during this period fingerprints were used to sign and validate contracts. Over the last century, biometrics -- the science for determining a person's identity by measuring his/her physiological characteristics -- has grown enormously. Technologies are being developed to verify or identify individuals based on measurements of the face, hand geometry, iris, retina, finger, ear, voice, speech, signature, lip motion, skin reflectance, DNA, and even body odor. In this course we will explore the latest advances in biometrics as well as the machine learning techniques behind them. Students will learn how these technologies work and how they are sometimes defeated. Grading will be based on homework assignments and a final project. There will be no midterm or final exam. Prerequisites: a background at the sophomore level in computer science, engineering, or like discipline.

 

 

Text

Pattern Classification, Duda, Hart, and Stork

Sections Covered: Chapter 1, 2.1—2.7, 3.1—3.8

Matlab, Student Version

 

 

F.A.Q.

Questions? Please check the F.A.Q.

 

 

Lectures

Introduction to Biometrics 

Introduction to Face Recogntion

Basic Probability + Introduction to Pattern Classification

Bayes Decision Theory

Notes on Matrix Differentiation

Whitening Transform Code

PCA and Eigenfaces

LDA and Fisherfaces

Decision Trees

Support Vector Machines

Hand Written Notes on Pattern Classification

Nonparametric Techniques

Plant Identification 

DNA Barcoding

Physics of Image Formation

Face Detection and Face Recognition

Face Recognition and Face Search

Iris Recognition     Daugman Paper     Industry Sales Pitch

Fingerprints

Face Detection

 

 

Topics

 

 

 

Assignments 

Turn in your assignments by dropping them in the mailbox marked Biometrics in the TA room on the first floor of Mudd (map).

Assignments marked as "(last year)" have not yet been updated for this year's class. Do not do them yet. They will change.

 

Assignment 1                           Due: October 4     

 

Assignment 2                           Due: October 11

 

Assignment 3                           Due: October 18 Trainfile   Testfile Matlab Primer Sample Matlab File

 

Assignment 4                           Due: October 25 Trainfile   Testfile  Handpoints

 

Assignment 5                           Due: November 1 November 6 Trainfile   Testfile

 

Assignment 6                           Due: November 15 November 20 Faces   Folds   matlab code: vlfeat_and_libsvm read_lfw_folds roc

 

Final Project            Project Description  +  Project Presentations  +  Final Project Report

 

 

Additional Datasets: PIE WITH 2 POSES     COMPLETE UNPROCESSED PIE

 

 

Resources

 

Face Detection

 

 

 

Datasets

 

PID

 

BCFD raw jpgs

 

BCFD aligned and cropped  (compressed version)

 

BCFD aligned and cropped  (high-resolution 16 bit)

 

 

Grading