New Course: CS-395/495 Section
IBMR: Image Based Modeling and Rendering
Recent computer graphics work promotes
digital images to a 'first class' primitive, to make images equally useful as
both input and output. In this course we will learn about:
Imaging: how can we 'stitch together' overlapping photos into one continuous
Mattes: A glass-and-silver goblet reflects and distorts its surroundings
in wonderful ways.
How can we make a photograph of that goblet behave in the same way?
Interpolation: Learn how to make the Mona Lisa (in Leonardo DaVinci's
painting) turn her head
from side to side. How can we change camera positions long after the photo
session is over?
Modeling: How can we make 3D views of complicated objects without 3D shape
Modeling: How a camera can measure the shape of a buttery croissant, a
tiger, a cloud, a waterfall, or a flickering candle.
Lighting: How can we use photographs as substitutes for light
sources in computer graphics renderings? Or in existing photographs? For
example: can we capture dappled warm light from the forest floor and use
it to light a computer-graphics dinosaur? Can we apply that lighting to a
photograph of Sue, the Tyrannosaurus Rex skeleton on display at the Field
Rendering: How can we make new computer graphics renderings from images
Spring Quarter 2003: Tues, Thurs 3:30-5:00pm, 1890 Maple Ave. Rm 342
Instructor: Jack Tumblin ( email@example.com , www.cs.northwestern.edu/~jet )
to Computer Graphics, reasonable comfort with linear algebra
or permission of instructor (please ask if you're
interested--you can probably do it!)
Grading: take-home mid-term,
take-home final, and one programming project built and graded at several
Projects will receive "Image-based grading"--let's make some