Instructors: Brian Curless and Steve Seitz

Times: Wed 3:30-5:20, Fri 3:30-4:20

Location: MGH 238

Description:

3D photography is the process of using cameras and light to capture the shape, appearance, and motion of real objects. This process provides a natural way of acquiring 3D models of unparalleled detail and realism by scanning them in from the real world. It is gaining wide use in digital film houses and industrial design studios and is a hot research topic in computer vision, computer graphics, and robotics.

In this course, we will cover topics that include the fundamentals of lenses and solid state sensors, passive vision for shape recovery, structured light scanning, BRDF capture, surface registration and reconstruction, smooth surface fitting, model construction and instrumentation for animation, and special topics related to emerging research areas in the field. The material will be a mix of lectures, discussion, and student presentations of project work. In addition to Brian and Steve, we will have a number of industrial and academic experts presenting cutting edge material in the realm of 3D photography.

The class will be project-oriented, starting with hands-on construction of real 3D photography systems. We will have a suite of controlled lighting sources (including a new DLP projector), digital still and video cameras, motion control platforms, and supporting software. The final projects will be research-oriented, intended to explore new areas of 3D photography and ultimately lead to quals projects and publications in leading vision, graphics, and robotics conferences.

Prerequisite: 557 or 590SS or equivalent

Enrollment: CSE grads or by permission of instructor