Stat 238

Vision as Bayesian Inference

Mon/Wed: 12:00-1:20 Winter 2010, Franz 2258A.
 
www.stat.ucla.edu/~yuille/Courses/UCLA/Stat_238/Stat_238.html.
 

---

Course Description

This course models vision as Bayesian Inference. It concentrates on visual tasks such as segmenting images, detecting objects in images, and recognizing objects. Its goal is to describe the state of the art techniques. The handouts consist of copies of the lecture notes and related papers.

Reading Material

• Course Notes and Handouts.
• Papers and Journal Articles.
• Instructor: Prof. Alan Yuille, yuille@stat.ucla.edu, 310-267-5383, office Math Sciences 8967.
• Office Hours: Wednesday 4.00-5.00 pm (8967 Math Sciences), or by appointment (email).

Grading Plan: 3 homework assignments (20% each). Term project (40%).

Tentative Schedule.

Lecture	Date	Topics	Reading Materials	Handwritten  Notes	Handouts
1	01-06	Introduction to the Course: Statistical Edge Detection	statistical_regions.pdf	Lecture1.pdf	chp1.pdf
2	01-08	Probability Distributions on Graphs: Basic Introduction	GYtics.pdf	Lecture2.pdf	chp2.pdf
3	01-13	MRF models : With and without hidden variables.	segmentation_overview.pdf	Lecture3.pdf	chp5.pdf
4	01-15	Free Energies, EM and Mean Field Theory. Steepest Descent, Discrete Iterative Algorithms	YuilleMicroBook.pdf	Lecture4.pdf
5	01-20	Belief Propagation: Dynamic Programming		Lecture5.pdf	chp9.pdf
6	01-22	MCMC: Brief description of MCMC		Lecture6.pdf	chp4.pdf
7	01-27	Hidden Markov Models and SCFG's: Learning and Inference by Dynamic Programming	chang_peng_2002_2.pdf	Lecture7.pdf	Lecture7.5.pdf	chp6.pdf
8	02-01	Maximum Likelihood and Discriminative Learning: Without hidden variables. ML versus AdaBoost		Lecture8.pdf	chp3.pdf
9	02-03	Maximum Likelihood and Discriminative Learning: With hidden variables. Structure ML.	yu_joachims_09a.pdf	Lecture9.pdf	tsochantaridis_etal_04a.pdf
10	02-08	Graphical Models of Objects Introduction + more structural learning.	NIPS2003_AA04.pdf	Lecture10.pdf	hand_cviu00J.pdf fergus03object.pdf
11	02-10	Hierarchical Object Models			Lecture11.pdf
12	02-17	Hierarchical Models (Cont).	HLLM08pami_Alan.pdf	Lecture12.pdf	notes10.pdf notes11.pdf
13	02-22	Hierarchical Models (Learning).	RCM_sum.pdf TPAMI-2007-03-0146-2.pdf	Lecture13.pdf	notes12.pdf
14	02-2	Deep Belief Networks Active Appearance Models	deepbelief.pdf	Lecture14.pdf
15	03-01	Active Basis Mode	http://www.stat.ucla.edu/~ywu/AB/ActiveBasisMarkII.html	ywu.pdf
16	03-03	Lighting Models		Lecture16.pdf	chp11.pdf
17	03-08	Jury Duty
18	03-10	Geometry of Vision: I	SzeliskiBook_20100227_draft.pdf	Lecture17.pdf
19	03-10	Geometry of Vision: II
		Final Project Due 20/March. Hand in to Prof. Yuile. Office or Mailbox