Computer Science 455
Instructor: R. P. Burton
Fifth Quiz
Name
_________________________________________ Score
____________/
1.
Which of the following projection combinations is NOT possible?
a.
parallel orthographic
b.
parallel oblique
c.
parallel perspective
d.
perspective orthographic
e.
perspective oblique
f.
(all are possible)
(c)
2.
Where is the center of projection in a perspective projection?
a.
on the screen
b.
at the eye
c.
infinitely far away in z
d.
there is no center of projection
(b)
3.
Where is the center of projection in a perspective projection?
a.
on the screen
b.
at the eye
c.a.infinitely far
away in z
d.c.
there is no center of projection (or, it is
infinitely far away)
(c)
4.
Every
projection discussed in class typically is (converted to) a _______ projection
before screen coordinates are determined.
a.
parallel orthographic
b.
parallel oblique
c.
parallel perspective
d.
perspective orthographic
e.
perspective oblique
(a)
5.
An isometric projection is _______.
a.
a parallel orthographic projection
b.
a parallel oblique projection
c.
a parallel perspective projection
d.
a perspective orthographic projection
e.
a perspective oblique projection
f.
a projection that does not fall into any of these
categories.
(a)
6.
A(n) ______ projection preserves lengths of lines
perpendicular to the projection plane.
a.
axonometric
b.
cavalier
c.
cabinet
(b)
7.
An accurate perspective projection is realized by a division by
a.
the distance from the viewplane
b.
the distance from the viewer
c.
an arbitrary, but uniform distance chosen for the entire scene
(b)
8.
What is the maximum number of principal
vanishing points in a scene composed of 3D objects?
a.
one
b.
two
c.
three
d.
more than three
(c)
9.
Suppose the origin of the viewing coordinate axis has been translated
so that it is coincident with the origin or the world coordinate axis. Suppose further that it has been transformed
so, like the world coordinate axis, it is a right-handed system. What is the maximum number of rotations
required to make the two coordinate systems coincident?
a.
one
b.
two
c.
three
d.
more than 3
(c)
10.
How is an oblique projection converted to an orthographic projection?
a.
by scaling
b.
by rotating
c.
by shearing
d.
(it can’t be done)
(c)
11.
Suppose you have a line from (0,0) to (2,3),
and you wish to align it with the y axis by shearing. What is the shearing factor in row 1, column
2 of the 3x3 transformation matrix?
a.
2/3
b.
–2/3
c.
3/2
d.
–3/2
e.
none of the above
(b)
12.
If you do an perspective-oblique projection
directly to the window, the contents of the window is the same as if you had
converted from perspective-oblique to parallel-orthographic and then projected
to the window.
a.
true
b.
false
(a)
13.
The exclusive purpose for
near and far planes is to permit cross sections to be taken.
a.
true
b.
false
(b)
14.
The following 2D clipping techniques extend to 3D with the exception of
____
a.
Cohen-Sutherland
b.
Liang-Barsky
c.
Nicholl-Lee-Nicholl
d.
Sutherland-Hodgman
e.
Weiler-Atherton
f.
(no exceptions here)
(c)
15.
A point is “inside” a perspective-oblique view volume if, when its
coordinates are substituted into the equation for each plane
bounding the perspective-oblique view volume, the result is < 0.
a.
true
b.
false
(a)
16.
For parallel-oblique projection, it is possible to find a matrix which
transforms an arbitrary scene all the way from world coordinates to the
viewport in physical device coordinates.
a.
true
b.
false
(a)
17.
The Roberts algorithm, discussed in class, is an
______-space algorithm.
a.
image
b.
object
c.
hybrid image/object
(b)
18.
Which category of algorithms lends itself
better to zooming in and zooming out?
a.
image space
b.
object space
(b)
19.
Back-face removal eliminates approximately half of the _______ in an
arbitrary scene.
a.
polygonal faces
b.
hidden faces
(a)
20.
In its simplest form, back-face removal requires
a.
substitution of the viewpoint measured in viewing coordinates into the
equation of the plane of the polygonal face
b.
substitution of the viewpoint measured in “any” coordinate system into
the equation of the plane of the polygonal face
c.
consideration of the normal to the polygonal face (in viewing
coordinates)
d.
consideration of the normal to the polygonal face (in “any”
coordinates)
e.
consideration of just one of the planar coefficients (in viewing
coordinates)
f.
consideration of just one of the planar coefficients (in “any”
coordinates)
(e)
21.
Which hidden surface algorithm is most likely to be in use if the
surfaces appear on the display in order from most distant to closest (to the
viewer)?
a.
depth-buffer
b.
scan-line
c.
depth-sort
d.
area subdivision
(c)
22.
Which hidden surface algorithm is most likely to be in use if entire
surfaces appear in no particular order?
a.
depth-buffer
b.
scan-line
c.
depth-sort
d.
area subdivision
(a)
23.
Which hidden surface algorithm is most likely to be in use if the
surfaces appear in order from the top of the screen to the bottom of the
screen?
a.
depth-buffer
b.
scan-line
c.
depth-sort
d.
area subdivision
(b)
24. What is the minimum size of a functional z-buffer for the z-buffer algorithm?
a. one “pixel”
b. one “scan line”
c. the same size as the frame buffer
(a)
25. Finding the depth of one point on a plane (for one pixel) involves substituting x and y into the planar equation and solving it for z. Finding the depth for an adjoining pixel
a. involves the same amount of calculation
b. involves the addition (or subtraction) of a constant
c. involves a multiplication
(b)
26. The scan-line hidden surface method works for all of the following EXCEPT
a. two overlapping polygons
b. more than two overlapping polygons
c. interpenetrating polygons
d. transparency
e. (no exceptions here)
(e)
27. Suppose surfaces have been ordered preliminarily on their maximum z values. Then, either (1) the two polygons do not overlap in z, (2)the two polygons do not overlap in x or y, (3) the projections of the two polygons do not overlap, or (4) the vertices of one polygon are all on one side of the plane of the other polygon
a. true
b. false
(b – interpenetration)
28. For the area-subdivision algorithm discussed in class, which of the following is too complex to scan-convert (hence requiring further subdivision)?
a. several polygons intersect the area, but one surrounds it
b. a single polygon intersects the area
c. no polygons intersect the area
d. (none of these is too complex, as discussed in class, to scan-convert without further subdivision)
(a)
29. Which has the greater negative impact on the efficiency of the area subdivision algorithm?
a. the number of subdivisions required to process the scene
b. the visual complexity of the scene
(b)
30. Which of the following is more efficient if used for hidden line elimination (i.e. the production of “wire frame” images) than for hidden surface elimination?
a. z-buffer
b. scan-line
c. painter’s algorithm
d. area-subdivision algorithm
e. (none are particularly more efficient when used for hidden-line elimination)
f. (all are MUCH more efficient when used for hidden-line elimination)
(e)
31.
Diffuse reflection is scattered light from
a.
point source lights
b.
ambient light
c.
both (a) and (b)
(c)
32.
Specular reflections result from
a.
point source light
b.
ambient light
c.
both (a) and (b)
(a)
33.
The intensity of reflected point source light is
a.
independent of the orientation of the reflecting surface
b.
related linearly to the angle between the surface normal and the angle
of the incoming light
c.
related linearly to the cosine of the angle between the surface normal
and the angle of the incoming light
(c)
34.
Pick the best statement:
a.
Each point light source in a scene (potentially) contributes to the
intensity of each point on each surface in the scene.
b.
Like ambient light, point source light (contributing to the intensity
of points on surfaces in a scene) is represented by a single term, independent
of the number of point light sources.
(a)
35.
The specular reflection model, as discussed in class, considers (pick the most complete answer)
a.
the intensity of the incident light
b.
(a) and the distance between the incident light source and the surface
c.
(b) and angle between the viewer and the reflected ray
d.
(c) and the angle between the incident ray and the surface normal
(c)
36.
Light striking a partially transparent, rectangular object (like a box)
other than perpendicular to a surface, ultimately exits the object
a.
along a path which is a linear extension of the path of the incident
light
b.
along a path which is parallel to a linear extension of the path of the
incident light
c.
along a path with a direction different from (a linear extension of)
the path of the incident light
(b)
37.
Which of the following is either ineffective or least practical for
producing the appearance of texture?
a.
altering the surface normal (as a function of position over the
surface)
b.
altering the coefficient of reflection (as a function of position over
the surface)
c.
using texture mapping methods
d.
micropolygonization of the surface
e.
(all are effective and roughly equivalently practical)
(d)
38.
Array pattern mapping, to achieve surface details, requires the
“receiving” surface to be both planar and rectangular.
a.
true
b.
false
(b)
39.
Suppose that a “shadow” is created (and subsequently painted into the
frame buffer) by considering the position of the light source (such as a spot
light at the top of a circus tent) , the position and
orientation of the intervening object (such as a clown), and the plane which
receives the shadow (such as the floor on which the clown is standing). What sort of projection could be used or
adapted to determine the shadow?
a.
parallel orthographic
b.
parallel oblique
c.
parallel perspective
d.
perspective orthographic
e.
perspective oblique
(e)
40. Consider a collection of nonintersecting cubes, with faces of a uniform color on all cubes. Suppose that these cubes are inside a sphere with point light sources at the north and south poles, and with four light sources equally spaced around the equator, such that each face of each cube receives light from at least one light source. (true or false) There can be no shadows in such a scene.
a. true
b. false
(b)
41. Suppose you have system in which each pixel is either on or off. To achieve more than two levels of intensity, you decide to work with “megapixels” which measure n x m. How many different levels of intensity can you achieve with your m x n “megapixels”?
a. still just two
b. the permutation of n x m things taken n x m at a time
c. the combination of m x n things taken n x m at a time
d. n x m
e.
(n x m) –1
f.
(n x m) +1
(f)
42.
Continuing the previous question, how many more levels of intensity can
be achieved if each original pixel has 4 levels of intensity instead of just 2?
a.
no change
b.
not more than twice as many
c.
not more than 3 times as many
d.
not more than four times as many
e.
more than four times as many
( c:
[(levels of intensity – 1) x n x m] + 1)
43.
Adding noise can create the appearance of smooth edges and can improve
overall appearance.
a.
true
b.
nonsense!
(a)
44.
Constant shading (of an arbitrary polygonal surface), while relatively
easy to compute, models virtually nothing from real-world experience.
a.
true
b.
false
(b)
45.
Suppose you are using Gouraud shading to determine intensities along a
single scan line which intersects several connected polygons. If you linearly vary the intensity from one
side of each polygon to the other side, all evidence of the polygon boundaries
disappears.
a.
true
b.
false
(b)
46.
Which of the following is used to produce Gouraud-shaded images?
a.
weighting of the vertex normal according to the area of the polygons
meeting at that vertex
b.
weighting of the vertex normal according to the distance from the
vertex to the center of each polygon
c.
cubic interpolation
d.
interpolation of normals down edges
e.
interpolation of normals across scan lines
f.
(none of the above)
(f)
47.
Where does Phong first compute shades?
a.
at each polygon
b.
at each vertex
c.
along each edge (at each edge/scan-line
intersection)
d.
along each scan-line (at each pixel)
e.
(Phong does not compute shades)
(d)
48.
An unfortunate shortcoming of Gouraud and Phong shading is their inability
to preserve edges. A collection of
polygons, representing a face, for example, either is smooth-shaded or it is
not. Hence, a face tends to look like a
death mask, with no apparent boundary between the cheek and the lip, for
example.
a.
true
b.
false
(b)
[JAG1]These seemed to be two valid ways of saying the same thing