DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING
UNIVERSITY OF CALIFORNIA, SAN DIEGO

Instructor: Serge Belongie, Associate Professor, EBU3B 4118. Office Hours: click here.

Teaching Assistants: Andrew Rabinovich and Vincent Rabaud. (Office Hours: W 3:00-4:00 pm CSE 4127)

Note: when emailing the instructor with questions about the class, please put “cse252c” in the subject line or use the following address:

sjb+cse252c

cs.ucsd.edu

Class section id for CSE252C: #602196. Lecture: TuTh 2-3:20pm, Room TBA. Temporary room assignment is EBU3B 4140 until further notice.

Class mailing list: https://csemail.ucsd.edu/mailman/listinfo/cse252c.

Topics to be Covered: Object detection and recognition; visual category recognition; dataset issues; feature detection and description; object models; statistical pattern recognition based methods.

Prerequisites: linear algebra, calculus, probability and statistics. This course makes extensive use of Matlab. Click here for information on Matlab. Assignments should be prepared using LaTeX. If you are not familiar with LaTeX, click here.

Homework:

• Assignment #1 [tex] [pdf]
• Assignment #2 [tex] [pdf]
• Assignment #3 [tex] [pdf] [make_pointset.m]
• Assignment #4 [tex] [pdf] [generate_data_1.m]
• End of Quarter Exam [pdf]

Lecture Topics, Readings and Scribe Notes:

Part I: Introductory Topics.
• Lecture 1 (Sep. 27): Problem overview, datasets, performance evaluation.
Rabaud et al. (2005), Ch. 2, Pinz (2006), Ch. 1-2, Mundy (2006), Ponce et al. (2006), Fawcett (2004)
• Lecture 2 (Oct. 2): Images as vectors, distance-based classification, clustering. [pdf]
Golub & Van Loan (1996), Sec. 2.2, Lewis (1995), Jain & Dubes (1988), Ch. 2, Tuzel & Meer (2002)
• Lecture 3 (Oct. 4): Working with distributions, histograms-of-X, comparing distributions. [pdf]
Rubner et al. (2001), Schiele & Crowley (2000), Swain & Ballard (1991), Porikli (2005), Topsøe (2000)
• Lecture 4 (Oct. 9): Efficient nearest neighbor methods. [pdf] guest lecturer: Lawrence Cayton
Frome & Malik (2006), Shakhnarovich et al. (2003)
Part II: Detectors, Descriptors, Features.
• Lecture 5 (Oct. 11): Basics of image processing. [pdf]
Farid (2001), Lindeberg and ter Haar Romeny (1994), Part I, Perona (1995), Freeman & Adelson (1991)
• Prof. Belongie out of town Oct. 15-19 for ICCV.
• Lecture 6 (Oct. 23): Interest point detection. cancelled due to fires
  Mikolajczyk and Schmid part I (2004)
• Lecture 7 (Oct. 25): Interest point description. cancelled due to fires
  Mikolajczyk and Schmid part II (2005)
• Lecture 8 (Oct. 30): Pairwise clustering, spectral graph theoretic grouping. [pdf]
  Shi & Malik (2000), Meila & Shi (2001), Wertheimer (1923)
Part III: Object Models.
• Lecture 9 (Nov. 1): Modeling and estimating geometric transformations. [pdf]
  Bookstein (1989), Feynman (1966), Girosi et al. (1995), Hertz et al. (1991)
• Lecture 10 (Nov. 6): The correspondence problem. [pdf]
  Scott & Longuet-Higgins (1991), Pilu (1997), Gold, Rangarajan et al. (1997), Fischler & Bolles (1981), Dawes (2005)
• Lecture 11 (Nov. 8): Recognition without correspondence; bags of features. guest lecturer: Andrew Rabinovich
Nowak & Triggs (2006), Csurka et al. (2004), Lowe (2004), Rabinovich et al. (2007)
• Lecture 12 (Nov. 13): Shape Matching. [pdf]
  Belongie et al. (2002), Icke (2004)
• Lecture 13 (Nov. 15): Constellation and Part-based Models. [pdf]
  Geman et al. (1997), Fergus (2007)
Part IV: Statistical Pattern Recognition.
• Lecture 14 (Nov. 20): Learning mixture distributions, the EM algorithm. [pdf]
  Bilmes (1998), Tomasi (2004)
• Nov. 22 is Thanksgiving Day.
• Lecture 15 (Nov. 27): Kernel PCA, Support Vector Machines. [pdf]
  Schölkopf et al. (1998), Schölkopf et al. (1999), Burges (1998)
• Lecture 16 (Nov. 29): Boosting, AdaBoost, Cascades. [pdf]
  Freund & Schapire (1999), Viola & Jones (2001)
• Lecture 17 (Dec. 4): Final Project Presentations I. [reports]
• Lecture 18 (Dec. 6): Final Project Presentations II. [reports]

Related Courses:

• CS 295T, Object Recognition, Spring 2007 Kristen Grauman, UT Austin.
• 16-721 Learning-Based Methods in Vision, Spring 2007 Alyosha Efros, CMU.
• 6.899 Learning and Inference in Vision, Spring 2002 Bill Freeman, MIT.
• CS 598, High-Level Recognition in Computer Vision, Spring 2007 Fei-Fei Li, Princeton.
• CS 294, Recognizing People, Objects and Actions, Spring 2004 Jitendra Malik, UC Berkeley.
• CMPT 882, Recognition Problems in Computer Vision, Fall 2006 Greg Mori, Simon Fraser U.
• 9.912, Scene Understanding Seminar, Fall 2006 Aude Oliva, MIT.
• EE/CNS/CS 148 - Selected Topics in Computational Vision: Visual Recognition, Spring 2006 Pietro Perona, Caltech.
• CIS 680-301, Vision and Learning, Spring 2005 Jianbo Shi, U.Penn.

Most recently updated on Sept. 21, 2007 by Serge Belongie.