BackReturn Home

Basic Perspective

Perspective is how things seem to look from our Point of View (POV). The Picture Plane (PP) is the entire scene, and is usually equivalent to the "format" (i.e.: the outer edge of our paper or canvas). It is what the viewer sees. A good example is to imagine drawing on a window with a dry-erase marker what you see outside. The window frame would be the format and the glass itself would be the PP. Parallax means that any sort of movement will change our POV, so we have to stay still! To counteract it, we can use a Fixed Eyehole. A Fixed Eyehole is something unmoving that we look through to continuously keep the same POV while drawing from life. It could be a ring hanging from a string attached to the ceiling, or a hole punched into a card that is mounted on a stick placed firmly into the ground.

The Ground Plane (GP) is an imaginary surface that we are standing upon, no matter what our POV is. The Station Point (SP) is our location on the GP. On Earth, the "horizon" is where the ground and sky seem to meet. There is always a Horizon Line (HL), a horizontal line that seems to divide our POV into "top" and "bottom". This is because we are always standing on an imaginary GP, even if we aren't standing on Earth! If we do happen to be standing on the surface of the Earth, and are not looking straight up or down, then the HL is the same as the Earth's horizon.

Eye Level is our height above the GP. It usually coincides with the HL. Objects at Eye Level we face directly, objects below our Eye Level we look down at, and objects above our Eye Level we look up at. Our Line of Sight is the direction that we are looking. For example, we can look straight ahead so that our Line of Sight is parallel to the GP, or it could be tilted up or down.

A Bird's-Eye View is when we are far away from the GP, looking down on a scene from up high. A Worm's-Eye View is when we are close to the GP, looking up at a scene from down low. Usually, if the HL is high on the format, we are looking down (e.g.: more of the ground is visible), and if the HL is low on the format, we are looking up (e.g.: more of the sky is visible). Again, the HL almost always coincides with our Eye Level.

Diminution is when things seem to get smaller the farther they are away from us. Overlap is when things that are closer to us seem to cover up things that are farther away. This seems to separate everything in the environment into three "layers". From closest to farthest away, they are Foreground, Midground, and Background.

Atmospheric Perspective is when far away things seem fuzzy and have less detail. [Clarity diminishes with distance because light is travelling through an atmosphere that has a tendency to scatter it.]

Parallel lines that are moving away from us seem to meet (or "converge") at some point off in the distance. This is called Convergence. Because of Diminution, if things are far away enough, they will become so small that they seem to disappear! Therefore, the point where parallel lines seem to meet is called a Vanishing Point (VP). All lines that are parallel to one another share the same VP, and all lines that are not parallel with the PP will have a VP!

All lines that are level with the GP have a VP on the HL. Any lines that are tilted (i.e.: are NOT level with the GP) have a VP above or below where they would normally be on the HL if they were level. The line that connects all of these VPs together is called a Vanishing Trace. It is perpendicular (i.e.: 90-degrees) from the HL.

How much of a scene is within our focus at any given moment is determined by a Cone of Vision. We can think of the tip of the Cone as corresponding to an eye, and the base of the Cone as how much of the scene we see through that eye. A wider base means that the Cone of Vision encompasses more of a scene. This width is usually measured as an angle. [Why an angle? The side-view of a Cone is a triangle. If one side of a triangle became longer or shorter, the angle between the other two sides would also change.]

Notice that the human eye only takes in a limited amount. Things on the periphery of our vision are not as clear as those things within the center of our vision. This effect is simulated by a 60-degree Cone of Vision. Things seem to become distorted outside of this range. For example, a camera with a wide angle lens (or "Fish-Eye") can take in a larger angle, but things also seem to curve, bend, or warp in order to fit.

Depth of Field determines which aspects of the Foreground, Midground, or Background are in focus (e.g.: a background can be blurry while focusing in on something up close). When a thing is close to our eye, Overlap can make something that is relatively small seem quite large. Likewise, when a thing is pointed towards us, Diminution can make something relatively long seem very narrow. This effect is called Foreshortening.

The idea of "1-Point Perspective", "2-Point Perspective", and so on, has to do with how many Vanishing Points are used to draw a single object from a given POV. This is mostly based upon how many sets of parallel lines that an object seems to have within that POV. In actuality, a single object within a scene could have as many Vanishing Points as there are parallel lines going off at different angles. Therefore, the concepts of 1-Point, 2-Point, etc. are simplified examples that show us how to handle Vanishing Points in general.

Usually, 1-Point is used for objects that are facing us directly, 2-Point is used for objects that are tilted or rotated away from us in some way, and 3-Point is used for more extreme angles (like Bird's-Eye and Worm's-Eye Views). The VP that appears when looking up is called the Zenith, and the VP that appears when looking down is called the Nadir.

Try to learn how to use 1-Point really well before moving on to the other two types.

Other Things That We Can Do

Linear Perspective is when parallels are treated as lines. This would include 1-Point, 2-Point, and 3-Point.

Curvilinear Perspective is when parallels are treated as curves. For example, there is 4-Point and 5-Point, which produce something similar to a Fish-Eye. There is also 6-Point, which can show everything around a person, like a "Panorama". If you also include what you would see if you look up and down, then it is "Omnidirectional". It makes it seem as if we are looking in every direction simultaneously! [For a neat example, check out how the artist Richard Termes paints scenes on spheres.]

We can increase the amount of a scene that we see by using more sets of converging parallel lines. However, the image also distorts more, especially when it is presented on a flat surface all at once.

In addition to Linear and Curvilinear Perspective, there is also Paraline Drawing. There is no Convergence whatsoever in a Paraline Drawing. Lines are true parallels throughout! This is used within Drafting (i.e.: drawing blueprints).

Drafting usually focuses upon three different ways of looking at thing:

• The Plan (or "Top View", made up of Width, X and Depth, Z)
• The Elevation (or "Front View", made up of Width, X and Height, Y)
• The Section (or "Side View", made up of Depth, Z and Height, Y)

These are called Draft Views. They are drawn to scale (i.e.: a fraction of the size that they would be in real life), so that one can take accurate measurements right off of the page! However, as accurate as blueprints might be, they can also seem "flat" and not give a very good idea of what it would actually look like.

Thankfully, we can use any two of these Draft Views at the same time to make a 2-Point Perspective drawing. This is called Mechanical Perspective. For example, an architect might combine a Plan and an Elevation drawing of a building to show what it would look like when viewed at a specific angle.

You might wonder, "Why not start with the 2-Point drawing instead?". This is because Convergence would throw the measurements off. Therefore, people often find it much easier to do some Draft Views first. This principle applies to coming up with things in your imagination as well. For example, someone designing a character might make a "model sheet" that shows what they look like from the front and side before trying to draw them in a "three-quarter view".

If something is drawn with 2-Point Perspective and the Vanishing Points are too close together, then it causes distortion. This is sometimes known as "Violent" or "Forced Perspective". It appears because the further that we are away from something, the less Convergence that its parallels seem to have. Therefore, if there is too much Convergence, it just looks distorted. Again, only so much of a scene can fit into our view without any curvature. Spreading out the Vanishing Points is equivalent to opening up the Cone of Vision and/or changing the Depth of Field.

This understanding can lead to all sorts of different POVs, and even help us to simulate what different types of camera lenses would look like! For example, zooming out so that what is in the Foreground is pushed farther into the Background is like changing the "focal length" of a camera lens. Focal length is measured in millimeters (e.g.: 200-mm, 50-mm, 28-mm, etc.). Generally, the higher the focal length (in mm) that we use, the more zoomed in it seems. Less of the scene is visible all around.

Using Perspective Grids

A Perspective Grid is a series of guidelines that gives us a framework for drawing in 1-point, 2-point, 3-point, etc. Because we are always on a GP, it is easy to imagine a space readily partitioned by a Perspective Grid.

What is the purpose of a Perspective Grid? It acts as a point of reference for gauging:
• Relative height or depth
• Relative distance and size
• Relative orientation

Drawing with a Viewfinder makes copying images easier for the same reasons: It gives a point of reference for gauging the characteristics of things in relation to an observer's viewpoint!

Perspective Tips

Use Thumbnails To Plan Perspective

In other words, figure out the proper perspective on a small scale, and then increase its size! This is especially useful when dealing with VPs that are far off of the page. Remember, angles and proportions can scale in size without the relationships changing.

Use Photographs To Study Perspective

• Where is the HL in the scene? Is the observer physically standing on a GP, and if so, what is their height or Eye Level from it? Estimate the height by seeing if the HL passes through the Eye Level of other people in the scene and/or objects that can help you gauge sizes (e.g.: doorways, cars, etc.).

• Is there a set of parallel lines which can be traced back to a VP? If not, what if you were to place imaginary bounding boxes around objects in the scene? Look for Diminution as things near the HL. Note that Convergence always occurs away from the observer! [Although, there are some stylistic exceptions, such as within Byzantine art.]

• Can you divide the scene into Foreground, Midground, and Background? Look for differing sizes in Overlapping objects and how some objects become hazy due to Atmospheric Perspective.

We sincerely hope that this information has helped you to understand Perspective, and created a good foundation that will make it easier for you to learn more.