CSE/EE586 Computer Vision II
Mathematical Tools for Computer Vision
CSE Department, Penn State University
Instructor: Robert Collins

Spring 2010 Schedule: Tues/Thurs 9:45-11:00AM, 371 Willard

Presentation Order

Course Introduction
Collins Intro to Course (Jan 12)   [slides]   [6 per page]
Paper Search/Reading/Writing/Vision Resources

Gaussian Mixture Models and Expectation Maximization
Lecture Notes
Collins The Gaussian Distribution (Jan 12, 14)   [scanned notes]
Collins GMM and EM, Part 1 (Jan19)   [slides]   [6 per page]
Collins GMM and EM, Part 2 (Jan21,Jan26)   [slides]   [6 per page]
Incremental Programming Assignments
Set of EM assignments
hand in on Feb 9 as part of the EM Project.
Reference Material
Intro to Gaussian Distribution , Bishop, PRML book
Estimating Gaussian Mixture Densities with EM - A Tutorial, Carlo Tomasi.
Mixture Models and EM , Bishop, PRML book
Old and New Matrix Algebra Useful for Statistics, Tom Minka.
Research Papers / Oral Presentations
Thursday, Jan 28
Collins Statistical Color Models with Applications to Skin Detection, Jones and Rehg
sample presentation on the Jones and Rehg paper.   [slides]   [6 per page]
sample critique on the Jones and Rehg paper.
I'll also present some recent work done here, using mixtures of Bernoulli distributions.
Tuesday, Feb 2
Sitichai Adaptive Background Mixture Models for Real-time Tracking, Stauffer and Grimson
related paper of interest: Understanding Background Mixture Models for Foreground Segmentation, Power and Shoonees
Jingchen Tracking Colour Objects using Adaptive Mixture Models, McKenna et.al.
Project proposal: 1 page summary of what you are doing for the EM project.
Thursday, Feb 4
Vairoj Blobworld: Image Segmentation using EM ..., Carson et.al.
Ankit Mixture Models for Optical Flow Computation, Jepson and Black
Critiques Due: written critiques for at least 2 papers we read.
Thursday, Feb 18
Spotlight Presentations: 2-3 minutes presentation of your EM project.
Projects Due: code + writeup + spotlight slides.
EM Project Ideas
- From Bishop book: use kmeans or EM to label/compress colors in an image. Try diff values of K. maybe compare k-means and EM results. Bishop's idea text is available here
- From Forsyth and Ponce: a number of ideas, including color/texture segmentation (e.g. blobworld), fitting line segments (maybe compare EM with RANSAC?), motion segmentation, background subtraction (Power and Shoonees paper). See the [Forsyth and Ponce ideas].
- Two other project ideas from the last time I taught this course. One involves motion segmentation (I provide a dataset), one involves experimenting with the Jones and Rehg skin color classifier. [here are the ideas].
- Of course, I encourage you to come up with your own project ideas involving research or data that is relevant to you.

Procrustes Analysis
Lecture Notes
Collins Intro to Procrustes Analysis (Feb 9,11)   [slides]   [6 per page]
Least Squares Approach. scanned notes for [translation] and [rotation]
Solving for 2D rot, trans, and scale using [complex numbers]
Collins Background Material on PCA (Feb 11)   [slides]   [6 per page]
Collins Active Shape Models (Feb 16)   [slides]   [6 per page]

Graphical Models
Lecture Notes
Collins Intro to Graphical Models (Feb 23,25)   [slides]   [6 per page]
Collins Hidden Markov Models (Mar 2)   [slides]   [6 per page]
Collins Kalman Filter derivation (Mar 4)   [scanned notes]   These are lecture notes from my tracking course.
Don't freak out when you see a reference to "homework" on the very first page!
Reference Material
PRML Graphical Models Chapter, Bishop, PRML book
HMM Reading: A Tutorial on Hidden Markov Models, Rabiner
Research Papers / Oral Presentations
Tuesday, Mar 16
Khoa A Hidden Markov Model Framework for Video Segmentation Using Audio and Image Features, Boreczky and Wilcox
Soumya Gait-based Recognition of Humans Using Continuous HMMs, Kale et.al.
Rachana Automatic Analysis of Multimodal Group Actions in Meetings, McCowan et.al.
Thursday, Mar 18
Masaru Applications of Hidden Markov Chains in Image Analysis, Aas et.al.
Nicholas Hidden Markov Models for Face Recognition, Nefian and Hayes
earlier paper by Samaria that the above paper refers to
Razvan Investigating HMMs' Capabilities in 2D Shape Classification , Bicego and Murino
Thursday, Mar 25
Critiques Due: written critiques for at least 2 papers we read.

Markov Random Fields (Graphical Models II)
Lecture Notes
Collins Brief Intro to MRF (March 25)   [scanned notes]
movie Dan Huttenlocher (March 30)   Speeding Up Belief Propagation
MRF Research Papers / Oral Presentations
Tuesday, Apr 6
Hung-Hsuan Interactive Graph Cuts fpr Optimal Boundary and Region Segmentation, Boykov and Jolly.
Anuradha Comparison of Graph Cuts with Belief Propagation for Stereo, Tappen and Freeman.
Jeonghyung An Application of Markov Random Fields to Range Sensing, Diebel and Thrun.
Thursday, Apr 8
Mi Sun Park Learning 3D Scene Structure from a Single Still Image, Saxena, Sun and Ng.
Bharath Consistent Segmentation for Optical Flow Estimation, Zitnick, Jojic and Kang.

Sampling and Markov Chain Monte Carlo
Lecture Notes
Collins Intro to Sampling (Apr 13)   [slides]   [6 per page]
Collins Markov Chain Monte Carlo (Apr 15)   [slides]   [6 per page]
Research Papers / Oral Presentations
Tuesday, Apr 20
Jang Young Monte Carlo Localization for Mobile Robots, Dellaert et.al.
Wen-Yu Sampling Plausible Solutions to Multi-body Constraint Problems , Chenney and Forsyth
Eric Human Upper Body Pose Estimation in Static Images, Lee and Cohen