From 2D to 3D: Analyzing how artists sketch
Summary of the study.
Photographs of the images produced by the study.
Examples of the types of strokes made by the artists.
Examples of the variation in the strokes.
Complete sketching sessions for the horse model.
Complete sketching sessions for the train model.
Complete sketching sessions for the teapot model.
Complete sketching sessions for the Mr. Potato Head model.
Ultimate goal: To develop a 3D computer-aided sketching
system that lets artists sketch a 3D shape using traditional 2D sketching
skills.
Study goals:
1) To better understand
how artists use sketching both to capture an existing 3D shape and to develop a
novel one.
2) To develop
storyboards for 3D shape sketching. Each storyboard will consist of a
"sketch" that defines a desired 3D shape. The sketch itself consists
of a sequence of 2D strokes from one (or more) viewpoints. These storyboards
will be used to guide the creation of a 3D computer-aided sketching system.
Questions to be answered by the study:
1) Is there a
consistent ordering as to how people add strokes? Front to back? Major to
minor?
2) What is the
decision making process about what strokes to add?
3) Are artists
thinking in 3D or thinking in a perspective projection?
4) How many viewpoints
are sufficient? Does each viewpoint need to be complete? Are some viewpoints
better than others?
5) Does the sketching
process differ if the artist is trying to depict an existing object versus
invent an imaginary one?
Study setup:
Each participant will
be asked to create two (or more, as time permits) complete set of drawings.
Each drawing set will consist of one or more viewpoints. We will not fix the
number of viewpoints ahead of time, but instead will ask the participants to
provide sufficient information for a sculptor/modeler to create a 3D copy of
the object from the drawings.
Set 1) Draw an object
that's placed in front of them (they're free to pick up the object and move it
around).
Set 2) Draw an
imaginary object of a similar type.
The drawing surface
will be video taped while the participants draw so that their sketches can be
captured. A hi-res photo will be taken of the finished drawing as well.
After finishing the
drawings, the participants will be asked to view the videos and narrate what
they were trying to accomplish. They will also be asked to discuss/critique
someone else's final sketch of the same object, and (time permitting) also
narrate over that participant's sketching process.
Things to comment on:
1)
Symmetry
2)
Order of strokes
3)
Which strokes are structural, which texture
Types of objects:
1) Architectural scene
(toy houses)
2) Characters (horses,
dragon)
3) Man-made object (fire
truck)
Expected time: Under two hours.
Will probably do one
object of each class, or a small scene with more than one object type (eg, a
character standing next to a chair).
Concerns:
1) Asking the
participants to record train-of-thought is likely to disrupt their natural
sketching, since sketching is a fairly cognitive activity. Therefore we will
narrate afterwards.
2) Range of
applicability: Sticking to objects that could plausibly be built/modeled from
example sketches. Man-made objects, buildings, creatures, people, simple
plants. NOT: trees, grassy fields, complex combinations of parts.