If you’ve ever visited a Disney theme
park, or if you happen to be a fan of B-rated films from the 1950’s, then
there’s a good chance that you have been tricked by the illusion of forced perspective;
A technique that employs visual and psychological cues to make objects appear larger, smaller,
closer, or farther away than they really are. Examples of forced perspective can be found just
about everywhere if you have a sharp enough eye, from ancient architecture in Greece and Rome,
to modern photography and blockbuster films. It is also used extensively at many
theme parks and tourist destinations, but especially at the Disney parks where
Imagineers use just about every trick in the book to increase the level of immersion.
Forced perspective allows them to build unbelievable structures that would otherwise
be impossible, or at least too impractical to construct in reality, and it’s a technique
that Disney has mastered over the past 65 years. There is definitely a fine art to
creating good forced perspective, however it is also something that is
heavily grounded in math and science. This is the part that I mainly want to focus on in
today’s video, and once we have an understanding of the underlying principles, we will then
use them to explain why some instances of forced perspective work so well at Disney,
and why others don’t really work so well. To start things off, we should begin by taking a
look at the different ways that humans perceive depth, which is something we have become
quite good at through many years of evolution. Like many mammals, humans have two front-facing
eyes that provide us with stereoscopic vision, meaning that we are able to view the
world in 3-dimensions by interpreting images from two slightly different angles.
When both eyes are focused on the same object, our brain can use the difference in
perspectives to estimate how far away it is by using a process called triangulation.
This is basically just a trigonometry problem where we can solve for the length of the
triangle using the distance between our eyes and the angle of each image, however our
brain does this subconsciously in real time. The ability to perceive depth in this way
is commonly referred to as stereopsis, and it allows us to estimate distances
with a high degree of accuracy. It is also the exact same concept that tricks our
brain into seeing depth when watching 3D movies, because the special glasses allow each
eye to view a slightly different image. Another way that humans perceive
depth is called convergence, and this is pretty similar to
stereopsis, however it is based on muscle sensations rather than visual cues.
When an object is placed relatively close to us, our eyes tend to converge inward so that
they can both focus on the exact same point. This movement is controlled by extraocular
muscles that are attached to each eye, and our brain can sense how much these
muscles have contracted or relaxed in order to determine the angle of convergence.
That angle can then be used to calculate exactly how far away the object is by using a
bit of trigonometry, again something that our brain handles subconsciously.
Depth perception by convergence only works well when the angle is relatively
large, however, and it becomes practically ineffective when an object is further than
30 ft because the angle approaches zero. This can actually be used to our advantage when
trying to create forced perspective, because we can limit the viewer’s ability to perceive depth
simply by keeping them more than 30 ft away. If you want to see just how much of our depth
perception comes from stereoscopic vision, all you have to do is to cover one eye and try
reaching around for a few objects in front of you. You should find that it is much more
difficult to judge how far away things are because your vision can no longer
rely on stereopsis and convergence. Fortunately, the human brain can also use
monocular cues in addition to binocular or stereoscopic cues, which allows us to perceive
depth even when one eye is not functioning. One way that we do this is through the
process of accommodation, which involves the contraction and relaxation of ciliary
muscles that control the shape of our eye lens. When we try to focus our eyes on objects that are
far away, the ciliary muscles relax in order to make the lens thinner and increase the focal
length, and when we try to focus on something up close, the muscles contact in order to make
the lens rounder and decrease the focal length. Our brain is able to sense the degree of
contraction, just like with convergence, and this provides us with an accurate
estimate of how far away the object is, as long as the distance is greater than 6 feet.
Another way that we perceive depth with monocular cues is based on the principle of angular size,
which simply refers to the physical size of the image that is projected onto our retina.
The angle that a given object spans on the retina is directly proportional to the
angle created by the light entering the lens, and it is dependent on both the actual size
of the object and its distance from our eye. This means that if we are already
familiar with the size of a given object, such as a tree, or a car, or anything else
that we might interact with on a regular basis, then our brain can use the angle subtended on
our retina to calculate how far away it is. This method of depth perception always relies
on pre-existing knowledge, or at least a visual reference, but as long as we have a rough
idea of either the scale or the distance, then our brain can estimate the
other simply using the angle. However, since the apparent size
of an object changes with distance, it is also possible to trick the brain by
manipulating angular size, which is the fundamental technique behind forced perspective.
By making a familiar object larger or smaller, it can appear closer or farther away since this
would have the same effect on its angular size, and conversely, moving the object closer or
farther can also make it appear larger or smaller. The relationship between size and distance
follows the rule of similar triangles here, so if the viewpoint is fixed, then we can
calculate exactly what the size and position should be in order to achieve the desired effect.
As long as the difference between the actual and perceived scale is subtle enough, then this
trick alone can be enough to make our brain think that it is real, even when all our other
methods of depth perception tell us otherwise. A classic example is Main Street USA at Disneyland
in California, where forced perspective has been used to make the buildings appear taller.
The first floor of each building is basically normal size at a scale of 9/10ths, however the
second floor has been built slightly shorter at a scale of 5/8ths, and when there’s a third
floor, it has been built at a scale of 1/2. Our brain is generally familiar with the height
of a building, and we have the first story of the façade as a visual reference, so our mind just
assumes that all 3 stories are the same size, and that the upper floors appear smaller
because they are farther away from us. This illusion can be pretty convincing if you’re
not actively looking for it, mostly because the difference in scale is relatively small, and the
buildings can only be viewed from the ground up. It is easiest to maintain forced perspective when
you can control the viewpoint and prevent guests from getting too close, otherwise the illusion
can be broken by revealing the true scale. This is something that is demonstrated
by the American Adventure pavilion over at EPCOT in Walt Disney World, where
forced perspective has been used to make the building appear smaller rather than larger.
The façade of the pavilion has over-sized features like windows and doors that make it appear
as a regular two and a half story building, and I have to admit that it can actually look
pretty realistic when there’s nobody around. The problem here is that guests are free to
walk right up to and inside the building, which gives us a visual
reference for its true size. Our brain is quite familiar with how tall
people are on average, so having a bunch of guests stand directly in front of the building
quickly breaks the illusion, revealing that the pavilion is actually 5 stories tall.
Now, not too far away is a perfect example that shows just how good forced perspective can
be when the viewpoint is totally controlled, and that’s the corridor scene in
Tower of Terror at Hollywood Studios. In addition to the really cool Pepper’s ghost
and lighting effects, the hallway itself is made to look much longer than it really is
by sloping the floor, walls, and ceiling. Because of angular size, parallel lines will
always appear to converge at a single vanishing point, and it’s possible to increase the perceived
distance by making the lines converge faster. Since the size of the hallway decreases linearly
along its length, and smaller objects tend to appear farther away, it creates the illusion
that the hallway is nearly 70 feet long, even though the sloped portion of the
physical set is only about 6 feet deep. Of course, this is only effective when
viewed from just the right distance and angle within a controlled environment,
which isn’t really possible for large outdoor structures that can be
seen from all around the park. Things like mountains and castles tend
to push the limits of forced perspective a lot further as well, sometimes making them
appear as much as 50 times their actual size. The physics of light starts to become very
important at this kind of scale, and simply manipulating the angular size and distance is
not enough to make the illusion believable. When we look at things that are very
far away, they often appear hazy and dull since the reflected light has to
travel much further to reach our eye. This is primarily due to water droplets and
other small particles that are suspended in the atmosphere, because they reduce
apparent contrast and color saturation by scattering photons as
they travel through the air. Water droplets tend to scatter blue and
violet wavelengths more than other colors on the visible spectrum, and so distant
objects can also appear to have a blue tint because more of these photons get directed
towards our eye from incoming sunlight. For the purpose of forced perspective, it
can be useful to replicate these natural phenomena artificially to make it seem like nearby
structures are actually far off in the distance. This technique is commonly referred to as aerial
perspective, and it can be illustrated quite well with a simple animation of a city skyline.
Even though this is just a flat 2-dimensional image, we can create a sense of depth by
using dark saturated colors in the foreground, and lighter desaturated colors
in the middle and background. Stacking the layers on top of each other helps
to further the effect because our brain uses overlapping to perceive relative depth, and
placing certain layers higher up makes them seem farther away because we perceive
them as being closer to the horizon. We could also choose to add a bit of de-focus blur
to give the buildings a slight hazy appearance, and adding some shadows helps
to enhance the 3D perspective. It’s important that shadows in the
foreground are darker and more defined, while shadows in the middle and background should
be lighter and softer due to less contrast. The final technique that really brings
the whole animation together is moving the layers at different speeds to simulate
motion parallax, which is the phenomenon that causes nearby objects to pass through
our field of view faster than distant ones. This is yet another method that can be
used to create an illusion of depth, and it was actually pioneered by Disney
with the help of a multi-plane camera in some of their classic animated
films like Bambi and Snow White. Now, I’m not aware of any physical examples of
motion parallax over at the Disney theme parks, but there are many attractions that use aerial
perspective techniques to appear very far away, one of my favorites being Expedition
Everest at the Animal Kingdom. The artisans and Imagineers made
excellent use of color here, starting with darker tones for the rockwork around
the base of the mountain, and then transitioning to lighter tones for the rockwork higher up.
The white snow also furthers the illusion of scale because we are already familiar with
the idea of giant snow-capped mountains, and it even has some hints of blue mixed in
to reduce the contrast against a blue sky. Overlapping the various peaks and staggering their
height helps to create some added depth as well, and certain faces have been painted with darker
colors in order to produce artificial shadows. If you look at the attraction from just the right
viewpoint where it is partially hidden behind the tree line, then it really does look like a
huge mountain range off in the distance, even though the actual height is just under 200 ft.
The only real problem with the forced perspective is that the lift hill goes right up into the
mountains, and so every time a train goes by, the illusion kind of gets thrown off by giving
us a visual reference for its true size. For comparison, the rockwork over in Cars Land
at Disney’s California Adventure uses many of the same techniques as Everest, however the
illusion is more effective because there’s nothing around to give away the scale.
The aerial perspective is also close to perfect in Cars Land, with Ornament Valley almost
having a hazy appearance way at the back, and in my opinion, this might be the very best example
of forced perspective in all the Disney parks. In addition to colors losing saturation
and brightness over large distances, textures also tend to lose detail
and sharpness, and this is something that was captured very well here with all the
different layers of rockwork at varying depths. Another land that makes really good
use of texture for forced perspective is Pandora back at the Animal Kingdom,
particularly with all the vegetation on the show buildings and floating mountains.
The entire area is filled with a variety of tropical plants that have a lot of color
fine details, but up on the mountains, they’ve mostly used plain mosses and leaves
which are meant to look like distant forests. Conceptually, the idea is that the mountains are
so high up that the forests appear very tiny and faint when viewed from the ground, however there
is some debate about whether this and the whole floating illusion are actually effective.
I think part of the problem is that we don’t have any prior knowledge about how
large the mountains are supposed to be, since they come from a fictional world, and
that makes it really difficult to trick our depth perception by manipulating angular size.
Combine this with the fact that guests can walk all the way around and even under the structure
to get an up-close look, and it’s fairly easy to see why the illusion gets lost for so many.
Personally, I can kind of see the forced perspective if I look at it
from just the right angle, but I’m curious to know, in the comments, whether
you can see it and how effective you think it is. It may not be Disney’s best implementation of the
illusion, but it’s certainly not the worst either. And now, of course this video would not
be complete without talking about the most iconic forced perspective attractions,
and those are the 7 famous castles that can be found at Disney parks all around the world.
Some of the castles look very similar to each other, while others are more unique, but they all
share many of the same characteristics that make them appear much taller than they really are.
Just like the buildings on Main Street USA in California, the first level is constructed at
a 1:1 scale, and the upper levels gradually decrease in size as they get higher and higher.
The bottom portion typically consists of giant stone walls and a grand archway
that are intended to be life-size, or at least as large as we might expect them to be
if we have never seen a real castle before, while the upper portion consists of tall slender spires
with lots of intricate architectural features. The various elements give the design an elaborate
and royal appearance, but they have also been carefully arranged so that it is difficult
to distinguish between individual floors. This helps to keep the height more
ambiguous because our mind can’t determine the scale by counting the number
of stories, which is especially important when viewing the castle from a distance.
The upper portion is also thinner than the bottom and set back slightly so it appears
farther away, and it is meant to look like the stone walls are actually positioned in front of
the castle to create an illusion of greater depth. I find that this effect is most apparent on the
original Sleeping Beauty Castle in Disneyland, where it almost seems like the pink section
is part of a separate structure on the other side of the wall, but in reality,
it’s all part of the same building. The contrast in color and texture also
helps to increase the perceived depth here, and you might notice that the tones gradually
get lighter from the bottom to the top, which furthers the illusion of height.
Many of the castles were originally painted with light pale colors in order
to achieve aerial perspective as well, but lately Disney has been going for more vibrant
colors that really stand out and catch the eye. Unfortunately, this does seem to have a slight
impact on the intended forced perspective, but the castles are so massive anyway that I don’t
think too much of the illusion is really lost. However, there is one other castle that
we still need to talk about which doesn’t have the advantage of scale, and that is
the Beast’s castle in the Magic Kingdom. This is probably the worst instance of forced
perspective that you can find at any of the Disney parks, but it’s a perfect example of what
happens when the illusion is pushed too far. The castle only stands about 20 ft tall
on top of the Be Our Guest restaurant, however it is supposed to look closer to
400 ft, which is 20 times its actual size. For comparison, Cinderella castle at the end of
Main Street is also supposed to look like 400 ft, but that’s only twice its actual size since it’s
already just under 200 ft tall to begin with. Forced perspective tends to work really well for
small to moderate changes in perceived scale, but it becomes less and less effective as
the scale stretched further and further. On paper, the concept of angular size tells us that the Beast’s castle and Cinderella
castle should both appear the same size, but in reality, it’s just not possible to fool
human depth perception with a miniature model. I tried to use Photoshop to make the
illusion look a little more realistic by lowering the saturation and contrast, and
adding in some blue tones with a bit of blur, but even the textures and level of detail are not
all that convincing from such a close viewpoint, and it loses any sense of aerial perspective
as soon as we go back to the original image. The castle simply appears too bright and
vivid with really harsh shadows, and the illusion is lost entirely as soon as you see a
person or a bird up there as a size reference. Aside from this, the Disney theme parks are still
home to some of the very best forced perspective illusions in the world, including so many
more that I couldn’t get to in just one video, and there’s really only a few where
something might seem a little off. I’d love to hear which ones are
your favorites down in the comments, as well as which examples you personally
find the most and least effective. I also want to say thanks to NordVPN
for generously sponsoring today’s video, which is a service that I like to use
for online security and IP masking. If you live in a country that has geo-restrictions
or government firewalls, then you can actually use a VPN to get around this by routing your
internet connection through a foreign server. That might sound pretty complicated, but
NordVPN makes it super easy with just a few clicks using their simple interface.
All you have to do is click on a country where you want your IP address to
appear, hit connect, and in seconds, you can browse the internet as if you and
your device were physically in that location. This is perfect for accessing online
video streaming platforms like Netflix, Disney Plus, and the BBC, which are only
available in specific regions, and NordVPN has no limit on bandwidth, so you’re free to watch
unlimited content from anywhere in the world. Of course, a VPN also provides an additional
layer of security at the same time by passing your traffic through an encrypted tunnel,
which prevents your ISP and third parties from tracking your online activities.
NordVPN uses military-grade encryption protocols with no data-logging, so your personal
information always remains private and secure whenever you are connected to their network.
I highly recommend that you give them a try by going to NordVPN.com/ArtofEngineering, where
they are currently offering a 2-year plan for 68% off with a 30-day money-back guarantee.
That works out to only $3.71 per month, and you can also enter promo code
ArtofEngineering to get one extra month for free. I have been extremely happy with their service
over the past 2 years, and they always help me out by sponsoring this channel, so definitely go
check them out at NordVPN.com/ArtofEngineering using the link in the description.
Don’t forget to leave a like and hit subscribe if you enjoyed today’s video,
and as always, thanks for watching!