Geometric Transformations: Definition, Types and Examples

Every frame of every video game is drawn by sliding, rotating, and reflecting triangles.

Not metaphorically. Literally.

The 3D world of a video game exists as a set of coordinates — vertices of triangles. The GPU reads those coordinates, applies a sequence of geometric transformations to each one, and projects the result onto your 2D screen sixty times per second. The dragon swooping from the left, the camera rotating to follow it, the character's shadow reflecting off the floor — every one of those effects is a transformation applied to a coordinate.

Geometric transformations are rules that move, flip, turn, or resize a shape while preserving its essential properties.

Geometric transformations are operations that change the position, orientation, or size of a figure on a coordinate plane. There are four main types: translation (sliding), reflection (flipping), rotation (turning), and dilation (scaling). The first three — translation, reflection, and rotation — are rigid transformations (also called isometries): they preserve both size and shape. Dilation changes size but preserves shape.

Different Types of Image Transformation Using Geometric Transformation

Translation — Sliding Without Turning

A translation moves every point of a figure the same distance in the same direction. The shape does not rotate or flip — it slides.

If you translate a point $(x, y)$ by $a$ units horizontally and $b$ units vertically, the new coordinates are:

$$(x, y) \to (x + a, , y + b)$$

A positive $a$ moves right; negative $a$ moves left. A positive $b$ moves up; negative $b$ moves down.

Worked example

Translate triangle $ABC$ with vertices $A(1, 2)$, $B(3, 2)$, $C(2, 4)$ by the rule $(x, y) \to (x + 4, y - 1)$.

1. $A(1, 2) \to A'(5, 1)$

2. $B(3, 2) \to B'(7, 1)$

3. $C(2, 4) \to C'(6, 3)$

The triangle shifts 4 right and 1 down. Size and shape are unchanged.

Reflection — Flipping Across A Line

A reflection flips a figure across a line called the line of reflection. Every point of the figure maps to a point on the opposite side of the line, at the same perpendicular distance from it.

Three most common reflection rules:

The reflected image and the original are congruent — same size, same shape. But if the original shape is not symmetric, the image will appear "flipped" or mirrored.

Worked example

Reflect point $P(3, -2)$ across the x-axis.

Using the rule $(x, y) \to (x, -y)$:

$$P(3, -2) \to P'(3, 2)$$

The x-coordinate stays the same; the y-coordinate changes sign. The point moves from below the x-axis to above it, at the same horizontal position.

Rotation — Turning Around A Centre Point

A rotation turns every point of a figure by a specific angle about a fixed point called the centre of rotation. The distance from each point to the centre stays the same — only the direction changes.

Positive angles rotate counterclockwise (CCW); negative angles rotate clockwise (CW).

The standard rotation rules about the origin $(0, 0)$:

Worked example

Rotate point $Q(4, 1)$ by 90° counterclockwise about the origin.

Using the rule $(x, y) \to (-y, x)$:

$$Q(4, 1) \to Q'(-1, 4)$$

The point has moved a quarter-turn counterclockwise. The distance from the origin, $\sqrt{4^2 + 1^2} = \sqrt{17}$, is unchanged.

Dilation — Scaling By A Factor

A dilation scales a figure by a scale factor $k$ from a fixed point called the centre of dilation. If $k > 1$, the figure enlarges. If $0 < k < 1$, it shrinks. If $k < 0$, the image also flips.

For dilation from the origin:

$$(x, y) \to (kx, , ky)$$

Unlike the other three transformations, dilation is not rigid — it changes the size of the figure. The image is similar to the original but not congruent (unless $k = 1$ or $k = -1$).

Worked example

Dilate point $R(2, 3)$ by scale factor $k = 3$ from the origin.

$$R(2, 3) \to R'(6, 9)$$

Every coordinate triples. The shape of the figure is preserved; its distances from the origin triple.

Why Geometric Transformations Exist — The Revolution Felix Klein Started

In the 1800s, mathematicians were in a crisis. New geometries were emerging — hyperbolic geometry, projective geometry, spherical geometry — and no one knew how they related to each other. Were they all versions of geometry? Were some "more valid" than others?

In 1872, a 23-year-old mathematician named Felix Klein stood up at the University of Erlangen and delivered a lecture that reshaped the field entirely.

His idea, which became known as the Erlangen Programme, was this: every geometry is defined by its transformations. Euclidean geometry is the study of properties that survive translation, rotation, and reflection. Projective geometry is the study of properties that survive projection. Spherical geometry is the study of properties that survive rigid motions on a sphere.

Suddenly, every geometry had a fingerprint — the group of transformations that left its essential properties unchanged. Klein did not just unify existing geometries; he gave mathematicians a way to discover new ones. Any consistent group of transformations defines a geometry.

The broader tradition runs deeper. Leonhard Euler (1707–1783, Switzerland) was among the first to formalise rotation in three dimensions — his theorem on rotation (1775) showed that any rotation in 3D space can be described by a single axis and angle. That result is the backbone of every 3D graphics engine today.

Common Mistakes With Geometric Transformations

Mistake 1: Rotating clockwise when counterclockwise is required (or vice versa)

Where it slips in: Any rotation problem where the direction of rotation is not explicitly drawn. Students often default to clockwise because clocks go clockwise, forgetting that positive angles in standard mathematics go counterclockwise.

Don't do this: Apply the 90° CCW rule $(x, y) \to (-y, x)$ when the problem specifies 90° CW — this gives the wrong image entirely.

The correct way: For 90° CW, the rule is $(x, y) \to (y, -x)$, not $(-y, x)$. A quick check: after a 90° CCW rotation, a point that was on the positive x-axis should move to the positive y-axis. If it moved the other way, the direction is wrong.

The second-guesser will rotate correctly, then doubt themselves and re-rotate — arriving at 180° from the original. Commit to the direction, then verify with one point before transforming the whole figure.

Mistake 2: Confusing which coordinate changes sign in a reflection

Where it slips in: Reflection across the x-axis versus the y-axis. Students mix up which one negates x and which one negates y.

Don't do this: When reflecting across the x-axis, write $(x, y) \to (-x, y)$ — that is the y-axis rule, applied to the wrong line.

The correct way: The axis you are reflecting across is the one that stays constant. Reflection across the x-axis: x stays the same, y changes sign → $(x, y) \to (x, -y)$. Reflection across the y-axis: y stays the same, x changes sign → $(x, y) \to (-x, y)$. A memory anchor: the axis you cross is the coordinate that flips.

Mistake 3: Assuming dilation is the same type of transformation as the other three

Where it slips in: When a question asks which transformations preserve congruence and a student includes dilation.

Don't do this: List dilation alongside translation, rotation, and reflection as a rigid transformation.

The correct way: Dilation is non-rigid — it preserves shape (angles) but not size (distances). The resulting image is similar to the pre-image, not congruent. Only translation, rotation, and reflection are rigid (isometric) transformations.

The real-world consequence of mixing up transformation rules: in 1998, NASA's Mars Climate Orbiter was lost because two engineering teams applied different unit conventions to the same navigational data — one sent commands in pound-force seconds, the other expected newton-seconds. The spacecraft was directed to a trajectory it could not survive. The underlying error is the same category of mistake as applying the wrong transformation rule: a correct procedure, applied in the wrong coordinate frame, produces a result that is entirely off.

Rigid vs. Non-Rigid Transformations — A Summary

At Bhanzu, transformations are introduced through animation and drag-based GeoGebra activities rather than coordinate rules first — students discover that rotation preserves size by watching a shape spin without changing, then receive the algebraic rule as a description of what they have already seen.

Next Steps

Start with translation — apply $(x, y) \to (x - 3, y + 5)$ to a triangle with vertices $(0, 0)$, $(4, 0)$, $(2, 3)$ and plot both the pre-image and image. Then try a 90° CCW rotation of the same triangle. If you get stuck on the rotation rule, come back to the worked example.

For a challenge: can you design a sequence of two transformations that returns a triangle to its original position? There are several answers — finding one will tell you something real about what these rules actually are.

Want your child to explore transformations through live GeoGebra sessions with a trainer? Try a free class.