CSE 558 (Spring 2006) – 3D Photography

3D Photography

Course Overview

3D photography is the process of capturing the shape, appearance, and motion of real objects using cameras and projected light. This process allows us to scan real-world objects and create 3D models of them with unparalleled detail and realism. 3D photography is quickly becoming popular in both digital film houses and industrial design studios, and is a hot research topic in computer vision, computer graphics, and robotics.

This year's course will focus on capturing the appearance of real world objects. This appearance is determined by an object's reflectance properties, which explain how it interacts with light. The class will be project-oriented, and students will work in groups to build their own 3D photography system. The lectures will discuss relevant topics and techniques (e.g., camera technology, geometric and photometric calibration, and 3D scanning) and provide an in-depth introduction to the areas of appearance capture and modeling. Students will discuss and develop projects together in class. To present a broader perspective, we will invite a number of guest lecturers to present cutting edge material in the realm of 3D photography.

Project teams will have access to a suite of controlled lighting sources and projectors, digital still and video cameras, motion control platforms, and supporting software. The final projects will be research-oriented and are intended to explore new areas of 3D photography.

Instructors

Michael Goesele + guest lecturers
Aseem Agarwala (TA)

Time and Location

Tuesday 12:00 - 1:20
Thursday 12:00 - 1:20
Location: CSE 503

Office Hours

By appointment. Please email Aseem or Michael to make one.

Prerequisite

CSE 557 or 576 or permission of instructor

Enrollment

CSE grads or by permission of instructor

If you are not a CSE student, please fill out this form and turn it into the CSE front desk in order to get a CSE account.

Email List

Please join the mailing list for important announcements.

© 2006 Michael Goesele and Aseem Agarwala
Department of Computer Science and Engineering | University of Washington