Assignment #1 - perspective
Due Thursday, January 22 (in class)
CS 48N - The Science of Art
Winter Quarter, 2009
Your first assignment is to write a 4-5 page double-spaced paper on one of the
topics listed in the first section below. Alternatively, you may do one of the
projects listed in the second section. You may also choose another topic if
you clear it with me in advance. (Remember that at least two of your four
assignments for the quarter must be written papers. I will give you a similar
set of choices between paper topics and project ideas for the next three
Papers must be computer-printed, not handwritten. They will be graded on
effort, content, and style. In writing your paper, I encourage you to make use
of the books listed in the course bibliography
(handout #2). I have copies of each of these books, which I am willing to lend
out. I also encourage you to make use of the Stanford libraries. Don't forget
to cite your sources, using footnotes and/or a list of references, as
appropriate. Finally, I am happy to meet with you, during office hours or by
appointment, to discuss your paper and to offer suggestions for additional
Remember also that you must present one of your four assignments during one of
the three scheduled Student Days (February 3, February 19, and March 10). Your
presentation is not intended to be a reading of your paper, but a proper talk.
In other words, it should be lively and should engage the class in discussion.
People are encouraged to volunteer for these presentations, rather than being
chosen by me. I suggest choosing the topic you would most like to talk about
in front of your peers, and choosing the first Student Day following the
assignment on that topic to give your talk. If by the time of the class
preceeding each Student Day, not enough people have volunteered to present on
that day, I'll need to start asking people.
Leonardo's discussion of perspective (pp. 49-69) contains a wealth of clever
observations, but it also many errors. We considered one such error in class -
his claim that rectangular walls of extended length will appear hexagonal when
viewed from a position opposite the wall (p. 63). What other errors can you
find? Include cases where Leonardo's observation is correct, but his
explanation is wrong. This happens several times in his discussion of how the
eye works. David Lindberg talks more about this issue in his book
Theories of Vision: From Al-Kindi to Kepler. (His coverage of
Leonardo begins on p. 154.) In your analysis, be careful not to misinterpret
Leonardo's prose, which is sometimes vague or ambiguous. Remember also that
his "natural perspective" is our "linear perspective".
Martin Kemp, in the first part of The Science of Art, analyzes
Massacio's use of perspective in the Trinity fresco from both a
technical standpoint and a programmatic standpoint. The former derives a 3D
model from the painting, while the latter explains how Massacio uses
perspective to differentiate the mortal and immortal worlds. Apply Kemp's
methods to analyze the use of perspective in a painting not covered in Kemp or
in class. Gardner's book might be a good sourcebook from which to choose.
Consider both technical and programmatic standpoints. I am not, expecting you
to make a complete geometric analysis of the painting's perspectival
construction, as Kemp does, although you are welcome to try.
Write a "pamphlet", addressed to your fellow Renaissance artists, telling them
why they should use the newly invented science of linear perspective
in their paintings. Feel free to appeal to the general zeitgeist of the
Renaissance, but also enumerate as many specific advantages of perspective as
you can think of - both technical and programmatic.
Philip Steadman, in a paper reprinted in The Artful Eye, tries to
prove that Vermeer used a camera obscura to generate the nearly
perfect perspectives of his paintings, as many art historians have suspected.
Part of his proof consists of building a physical replica of the room in which
many of Vermeer's paintings are set. Read Steadman's account and make a
critical evaluation of it. Do you think he is right? Why or why not?
David Hockney, in his 2001 book
Secret Knowledge: Rediscovering the Lost Techniques of the Old Masters
advances the startling theory that many Renaissance artists, including van
Eyck, Holbein the Younger, and Caravaggio, employed optical devices such as
concave mirrors to project images onto their canvases, which they then traced
or painted over. If correct, this theory completely changes our interpretation
of these artists - indeed of a whole period in Western art. However, the
theory has been strongly criticized by art historians. An extensive web site,
giving both sides of the controversy, is available at
http://webexhibits.org/hockneyoptics/. Read Hockney's book and the web
site, then make your own critical evaluation of the issues. (One of Hockney's
most vocal critics is Dr. David Stork, a local researcher and occasional
lecturer at Stanford. Since Dr. Stork is local, he might consent to an
Kemp defines anamorphic images as perspectives that "assume undistorted
appearance only when viewed from a particular position" (p. 110). Anamorphic
images have gone in and out of fashion several times during the history of
Western art. Research and write a paper about this fascinating topic. Begin
with Kemp, p. 208-212, but look for other sources.
The making of curved or hybrid 2D-3D dioramas for entertainment and museums has
an interesting history. Diorama-makers employ many tricks we have not
discussed in class to enhance the realism of their illusions. Research the
history of dioramas and write a summary of your findings.
Multi-point perspectives, introduced during the Renaissance to reduce
distortion in large paintings (like Raphael's School of Athens), have been used
by artists for many purposes. 20th-century artists and cartoon animators have
introduced new uses for them, and recent papers in the computer graphics
research literature have proposed yet more uses. Write a paper about
multi-point perspectives, focusing on one or more of these uses or surveying
the entire subject.
Research and write a paper about a non-planar perspective presentation system.
Examples are Cinerama, Imax and Omnimax movie formats, planetarium domes (those
giant Zeiss projectors), panoramic images and movies on the web (some of these
use the QuickTime VR format), and so on.
Trace the historical development of projection systems, whether linear
perspective or not, of a non-Western-European culture. This will require some
Some non-writing projects
Make an accurate perspective drawing of a reasonably complicated
three-dimensional scene. Use a "correct" linear perspective construction
method such as the one presented in class. For this project, you will need to
know key dimensions of your scene in both plan and elevation, so choose your
scene carefully. A furnished but uncluttered bedroom has about the right
amount of complexity. Memorial Court may also be suitable, but measuring its
elevation (vertical dimensions) will be challenging; perhaps you can find
architectural drawings of it - in a library or through the Stanford
administration's Planning Department. Your submitted drawing should show your
station point, vanishing points if you used them, and all working lines (drawn
lightly). Draw neatly and use a straightedge, at least for your working lines.
Reproduce one of Brunneleschi's panels, now lost, in which he
demonstrated to his amazed contemporaries his ability to construct correct
linear perspective views. Kemp's book tells you how. Include the peephole and
mirror. Since the Baptistry of Florence is not nearby, choose another building
as your subject.
Model and render the 3D scene corresponding to a famous painting. Use a
commercially available modeling and rendering system such as 3D Studio Max or a
freeware/shareware modeling program you find on the web. For an extra kick,
script, render, and record on video a flyaround of the scene. Alternatively,
convert it into a form that can be navigated interactively on a PC (or Mac).
Beware: many paintings do not have a single, consistent 3D interpretation!
Build and demonstrate in class one of the perspective machines described in
Chapter 4 of Kemp. Choose wisely. An artist's glass (as in figure 330) is too
easy, and I already showed you one in class. Cigoli's machine (figure 351)
might be too hard. I'd love to see a demonstration of the machines in figures
331 or 332. The pantograph in figure 355 also looks interesting. Be
forewarned that unless your device is sturdily built, it won't be stiff enough
to make a reasonable drawing.
Copyright © 2009 Marc Levoy
January 15, 2009 10:44:24 AM