CS 638-1: Computational Photography, University of Wisconsin

Instructor: Chuck Dyer

  Office:       6379 Computer Sciences Building
  Telephone:    262-1965
  E-mail:       dyer@cs.wisc.edu
  Office Hours: Mondays and Wednesdays, 2:00 - 3:00 p.m., and by appointment

Teaching Assistant: Yancan Huang

  Office:       5388 Computer Sciences Building
  Telephone:    262-2542
  E-mail:       hyc@cs.wisc.edu
  Office Hours: Tuesdays and Wednesdays, 4:00 - 5:00 p.m., and by appointment

Schedule

Lecture: 11:00 a.m. - 11:50 a.m. MWF, Room 115 Psychology

Prerequisite

CS 367 or programming experience

Textbook

Computer Vision: Algorithms and Applications, R. Szeliski, latest online draft, 2009

Course Description

We are in the early years of an explosive growth of digital images. A recent study estimated that more than 500 billion images will be captured by digital cameras and cell phones by 2010. Because digital cameras allow easy capture of many images, billions of images are publicly available on the web, and computer storage and processing of digital images is cheap and easy, there is now emerging a wide range of new computational techniques and applications for capturing, analyzing, manipulating, combining, searching, synthesizing, and using images. Computational Photography is a new field that brings together photography, computer vision, and computer graphics to overcome the limitations of traditional cameras by computing new kinds of images and other representations of our visual world from sets of images. For example, Microsoft's Photosynth application allows users to interactively navigate around a 3D location by building a sparse 3D model from a large number of images. Key component image manipulation methods include warping, morphing, mosaicing, texture synthesis, segmentation, high dynamic range imaging, image blending and compositing, merging multiple exposures taken under different lighting conditions into a single perfectly exposed image, and building 3D models from a set of images taken from multiple viewpoints of an object or scene.

Grading

Homework assignments: about 50%
Course project: about 30%
Course project presentation: about 10%
Class participation: about 10%
Note: No midterm or final exam

Homework Assignments

Homework assignments will include written problems, using applications software, and programming in Matlab. There are many online resources for learning Matlab (e.g., see Getting Started), but if you want to buy a book, the following one is a good, simple introduction: Matlab Primer, 7th ed., T.A. Davis and K. Sigmon, Chapman and Hall Publishers, 2004. Another good alternative introductory book is: Getting Started with Matlab 7, R. Pratap, Oxford University Press, 2005. Accounts will be provided on the Computer Science Department's instructional Linux and Windows workstations. Assignment grading questions must be resolved with the instructor within one week after it is returned.

Late Penalties

All assignments are due in class on the due date. One (1) day late, defined as a 24-hour period from noon to noon the next day (weekday or weekend), will result in 10% of the total points for the assignment deducted. So, for example, if an assignment is due on a Wednesday and it is handed in between Wednesday 11 a.m. and Thursday 11 a.m., a 10% penalty will be deducted. Two (2) days late, 25% off; three (3) days late, 50% off. No homework can be turned in more than three (3) days late. A total of two (2) free late days may be used throughout the semester without penalty.

Academic Integrity

All examinations, programming assignments, and written homeworks must be done individually unless explicitly instructed otherwise. Cheating and plagiarism will be dealt with in accordance with University procedures (see the UW Academic Misconduct Policies and Procedures). Hence, for example, code for programming assignments must not be developed in groups, nor should code be shared, and code should not be obtained from anyone or anywhere, including the Web. You are encouraged to discuss with your peers, the TA or the instructor ideas, approaches and techniques broadly, but not at a level of detail where specific implementation issues are described by anyone. If you have any questions on this, ask the instructor before you act.

CS 638-1: Computational Photography