Skip to content →

Free Anamorphic Perspective Grid Generator

Plan anamorphic perspective art instantly with our 100% free grid generator. Set the viewing position, depth and skew, and the stretched grid shows exactly what to chalk so the illusion reads from one spot, then export SVG or 600 DPI in seconds. No signup, no downloads needed. Trusted by 60,000+ artists worldwide.

A4700 × 9000 overlaysSaved ✓

A grid stretched to be seen from one place

A deliberately distorted ruling that reads as correct only from a chosen viewpoint. Drawing into its cells produces an image that resolves from that spot and collapses from anywhere else — the pavement-art construction.

6 controls, not a fixed picture

Set columns, rows, viewer X and depth stretch, toggle construction rays to viewer point. 6 controls in all, every one live on the canvas.

Stacks with perspective & structure

Overlays combine, each carrying its own transform, so this one can sit under or over two-point perspective, isometric and 1-point perspective, and 4 more in the same set.

Print it at the size it will be drawn

The distortion is geometric, so it has to survive scaling: export as SVG for large-format printing or plotting, or at 600 DPI for tiled sheets.

The grid for pavement art, floor murals and every image built to snap into shape from a single viewing position — with the viewer point, the depth and the skew all in your hands.

The anamorphic grid generator draws one of the 82 composition overlays in Grid Maker Pro, and the rest stack on the same image.

  • 1 correct viewpoint
  • 4.1:1 default depth stretch
  • SVG + 600 DPI export
  • Nothing uploaded

What is an anamorphic perspective?

The anamorphic grid generator is a free browser tool that draws a stretched projection grid. The lattice is distorted so that, from one specific viewing position, it reads as a regular grid — and as nonsense from everywhere else. The viewer point, depth stretch and oblique skew are all adjustable, with optional construction rays. It draws over your own photograph or a blank canvas, exports SVG and raster to 600 DPI, and never uploads anything.

Anamorphic grid overlay on a long stretched light reflection on wet night pavement
Anamorphosis distorts a figure so it resolves from one place and one place only. A reflection on wet ground does the same thing without being asked: step sideways and the shape you were looking at is gone.
ProcessAnamorphic perspective grid generator

How to use the anamorphic grid

A grid stretched for a viewer standing where you say they will stand, so a drawing made on the floor stands up correctly from that one position.

  1. Decide where the viewer will stand

    Anamorphic perspective starts from the viewer's position, which is the whole design decision. Move the viewer point across the frame to match where people will actually see the finished piece from — a doorway, the end of a corridor, a marked spot on the pavement.

    JPG · PNG · GIF · WEBP

  2. Set the depth stretch

    Depth controls how hard the far end compresses. At 1.10 the grid is barely distorted; at 4.00 the nearest band is more than eleven times the height of the farthest. Match it to how low and how close the viewer's eye will be.

    Columns · Rows · Viewer X

  3. Turn on the rays, then export

    The dashed construction rays run back to the viewer point and are the fastest check that your setup is coherent. Export SVG for vector or a raster up to 600 DPI, then scale it to the real surface and transfer square by square.

    PNG · JPG · PDF · SVG · up to 600 DPI

Three steps, about a minute. No signup, no upload.

Capabilities

What the Anamorphic Grid Generator does

Compresses with distance

Anamorphic perspective shortens its rows as they recede, on a curve rather than linearly. At the default depth of 2.20 the nearest band is just over four times the height of the farthest — that ratio is what makes the flat surface read as receding ground.

One viewer point, movable

The point everything is built for. Slide it left or right and the entire construction re-aims. Placing it correctly against the real site is the difference between an illusion and a smear.

Oblique skew

For viewing positions that are not straight on. Skew leans the whole lattice up to 45% either way, which covers the common case of a piece seen from a walkway to one side rather than from directly in front.

Columns converge, rows compress

Two effects at once, and they are independent. Columns narrow toward the far edge — the near edge is about 2.3 times wider. Rows compress on the depth curve. Both are visible in the grid before you draw anything.

Over a photo of the site

Photograph the actual surface from the actual viewing position, load it, and lay the grid over it. Now the distortion is fitted to the real geometry instead of to an assumed one.

Vector out

SVG scales to any size without loss, which matters when the finished piece is six metres long. Raster export runs to 600 DPI, sized from the paper format rather than from your screen.

Plates

Anamorphic perspective examples: four ways the anamorphic grid reads

The default: viewer centred, depth 2.20, rays off.
The construction rays on, fanning back to the viewer point.
Oblique skew applied, for a viewing position off to one side.
High depth stretch, for a viewer standing close and looking down.
Comparison

Anamorphic perspective vs constructing it by hand from Nicéron vs a projector on site

OptionViewer point adjustableDepth controlVector exportCost
Grid Maker Prothis toolYesacross the full width1.10–4.00, plus oblique skewSVG + raster to 600 DPIFree
Constructing it by hand from NicéronYesYesn/aTime
A projector on siteFixed by the projectorPhysicalNoEquipment
3D software with a camera matchYesYesYesSubscription
The method

How an anamorphic perspective works — Anamorphosis: the oldest trick in perspective, and how the grid does it

Anamorphosis is perspective pushed until it breaks in a controlled direction, which makes it the one of the 82 composition overlays that is wrong from every position but one. The image is deliberately wrong from every position except one, and from that one it snaps into a solidity that an ordinary picture cannot reach.

Every picture made with linear perspective has a correct viewing position — the point from which the construction was projected. Stand there and the geometry is exact. Stand elsewhere and it is wrong, but only slightly, because the distortion grows gently as you move.

Anamorphosis takes the same principle and pushes the projection so far that the tolerance disappears. The image is built for one position, and from anywhere else it is not a slightly-off picture but an unreadable stretch. That severity is the effect, not a side-effect.

Where anamorphosis came from

The oldest surviving worked example is in Leonardo's Codex Atlanticus, around 1485: a drawing of an eye and a child's head, elongated so violently along one axis that they resolve only when the sheet is viewed from the right-hand edge at a shallow angle. It is a technical demonstration rather than a picture, which tells you the geometry was understood well before anyone thought of using it for a subject.

The famous case is Hans Holbein the Younger's The Ambassadors, 1533. Across the bottom of an otherwise conventional double portrait lies a pale diagonal smear. Move to the right of the canvas and look back along its surface, and the smear resolves into a human skull. The picture contains a memento mori that only appears when you leave the position the rest of the painting was built for.

That structural joke — the portrait works from the front, the skull from the side, and never both at once — is why the painting is the standard reference for anamorphosis. It is not a trick appended to the picture; it is a second picture occupying the same surface with a different viewing rule.

Seventeenth-century Europe produced a great deal of the same thing, much of it in Jesuit contexts, where the mechanics of projection were studied seriously. Emmanuel Maignan's 1642 fresco in Trinità dei Monti in Rome runs the length of a corridor and resolves into Saint Francis of Paola only from one end of it. Jean-François Nicéron's La Perspective curieuse, 1638, is the manual that documented how to construct these things.

How the distorted grid works

The mechanism in all of them is a grid. You draw the image on a regular square grid, then transfer it cell by cell onto a distorted grid whose cells correspond to the same squares as seen from the intended position. The transfer is mechanical; the projection is where the thinking goes.

This overlay is that distorted grid. Two things happen to the lattice, and they are independent.

A long raking shadow stretched far across pale paving by a very low sun
A shadow is a projection onto a surface at an angle, and it stretches exactly as far as that angle demands. Anamorphosis is the same distortion applied deliberately and then read back from one place.

The columns converge. The far edge of the grid spans a narrower band than the near edge — a ratio of about 2.3 to 1 at the default — so vertical lines lean inward as they recede. That is the same convergence any perspective drawing has.

What the depth control does

The rows compress, and this is where the depth control acts. Row spacing follows a curve rather than a straight line, so bands get shorter as they recede. At the default depth of 2.20, the nearest band is 4.09 times the height of the farthest. Turn depth down to 1.10 and that ratio falls to 1.19 — nearly uniform. Turn it up to 4.00 and it rises to 11.4.

Those three numbers are worth having in mind, because depth is the control people misjudge. It corresponds physically to how low and how close the viewer's eye is: a viewer standing right at the edge of a floor piece and looking down needs a high value; a viewer at the far end of a hall needs a low one.

Getting depth wrong produces a specific, recognisable failure. Too low and the piece looks like a picture lying on the ground rather than a hole in it. Too high, and from the intended viewing point the near end reads as absurdly enlarged and the illusion snaps the other way into a distorted drawing.

The viewer point and the oblique skew

The viewer point runs across the full width of the frame and answers a different question: where, laterally, the viewer stands. For a piece on a pavement with a marked photo spot, this should sit where the mark is. For a corridor, it is wherever people enter.

Oblique skew handles the case where the viewing position is not on the centre line at all — a piece seen from a walkway running along one side. Skew leans the whole construction up to 45% either way. Combined with the viewer point it covers most real installations, which are rarely as symmetrical as the diagrams in books.

The construction rays are dashed lines running from the viewer point back to the grid, and they are the fastest way to see whether your setup is coherent. If the rays fan evenly, the projection is behaving. If they bunch or cross oddly, the viewer point and the depth are fighting each other and the illusion will not hold.

How to lay out a pavement piece

For pavement and floor work the practical sequence is: mark the viewing spot on the real site first, photograph the surface from exactly there, load the photograph, and fit the grid to what the camera saw. That way the distortion is fitted to the real geometry — the actual slope of the ground, the actual eye height — instead of to an idealised version of it.

Then transfer. Draw your source image on the regular grid the rectangular grid maker gives you, at the same column and row count, and copy cell by cell. The mural scaling calculator works out what one cell measures on the ground. This is the same procedure as the ordinary grid method, and it is deliberately mechanical: the projection has already been solved by the grid, so the transfer needs accuracy rather than judgement.

Column and row counts run from four to thirty each. For a given amount of care, more cells means a more accurate transfer and slower work. For a large floor piece with organic subject matter, twenty or more is usual. For a hard-edged graphic, eight can be enough.

What this construction cannot do

There is a limit worth being honest about. This construction produces a plausible oblique anamorphosis for a flat surface seen from a stated position. It is a practical drawing aid, not a photogrammetric solution. If your surface is curved, stepped or on a slope, the grid gets you most of the way and your eye finishes the job — which is what pavement artists have always done.

The contemporary form of the tradition is street 3D chalk art, and its practitioners work almost exactly as Nicéron described, with the addition that they now check the illusion by photographing it from the spot as they go. The camera is a better judge than the eye, because the eye at the correct position corrects for what it knows about the ground.

The same geometry turns up in stage design, where a set is built to read correctly from the centre of the stalls; in trompe-l'œil ceiling painting, where the viewing position is the middle of the room; and in the road markings that read as upright text from an approaching car and as elongated smears from a footbridge.

That last one is the most common anamorphosis most people see daily, and it makes the principle concrete: the letters painted on a bus lane are wrong from every position except the driver's, and being wrong from everywhere else is what makes them right from there.

For export, remember that the piece is usually much larger than the sheet. SVG scales without loss, which matters at six metres. If you are printing a transfer sheet, export raster at your paper size and 600 DPI, then enlarge on site by counting cells rather than by scaling the print. The anamorphic overlay page works through the painted cases, beside the rest of the perspective overlays.

Everything runs in your browser. The photograph of your site is never uploaded, because there is no server for it to reach — the drawing, the transform and the export all happen on your own machine.

Studio notes

Where an anamorphic perspective gets used

A cinema projection booth with a carbon-arc projector beside a dark rectangular port window
A projected image is stretched by the angle it meets the screen at, and a booth is built off-axis by necessity. Correcting for that distortion is the whole trade.
Street artist3D pavement work

I photograph the pavement from the marked photo spot, load it, and fit the grid to that. Anything else and the illusion is off by the amount the ground slopes.

Mural painterfloor commissions

Depth is the control people get wrong. Too low and it looks like a picture on the floor instead of a hole in it.

Scenic designertheatre

For a set, the correct position is the centre of the stalls and everything is built for it. Same geometry, older problem.

Questions

Anamorphic perspective: frequently asked questions

The geometry

What is anamorphosis?

A projection built for one specific viewing position, pushed far enough that the image is unreadable from anywhere else. Every perspective picture has a correct viewpoint; anamorphosis removes the tolerance around it, so the picture snaps into shape at one place and dissolves as you move.

What is the oblique skew for?

Viewing positions that are not on the centre line. A piece seen from a walkway running along one side needs the whole lattice leaned over, and skew does that by up to 45% either way. Most real installations are less symmetrical than the diagrams in the books.

What are the dashed rays?

Construction lines running from the viewer point back to the grid. They are the quickest check that the setup is coherent: an even fan means the projection is behaving, and rays that bunch or cross oddly mean the viewer point and the depth are fighting each other.

Will this work on a sloping or curved surface?

Partly. The construction assumes a flat surface seen from a stated position, so on a slope or a curve it gets you most of the way and your eye finishes the job — which is what pavement artists have always done. Photographing the real surface from the real viewing spot and fitting the grid to that photograph closes most of the gap.

Where does the technique come from?

Leonardo's Codex Atlanticus has a worked example from around 1485. Holbein's The Ambassadors of 1533 is the famous painting. Jean-François Nicéron's La Perspective curieuse of 1638 is the manual that set out how to construct one, and street 3D chalk artists still work essentially by his method.

Using it

What does the depth control actually change?

How hard the rows compress as they recede. At the minimum of 1.10 the nearest band is 1.19 times the height of the farthest — nearly uniform. At the default of 2.20 the ratio is 4.09 to 1. At the maximum of 4.00 it is 11.4 to 1. Physically it corresponds to how low and how close the viewer's eye is.

How do I choose the viewer point?

From the real site, not from the drawing. Wherever people will actually stand — the marked photo spot on a pavement, the doorway of a corridor, the entrance to a hall. That decision comes before every other one, because everything else is built to serve it.

How do I transfer a drawing onto it?

The same way as the ordinary grid method. Draw the source on a regular grid with the same column and row counts, then copy cell by cell onto the distorted one. The projection has already been solved by the grid, so the transfer needs accuracy rather than judgement.

How many cells should I use?

Columns and rows run from four to thirty each. Twenty or more is usual for a large floor piece with organic subject matter; eight can be enough for a hard-edged graphic. For a given amount of care, more cells means a more accurate transfer and slower work.

Export and print

What do I print to take to the wall?

The grid alone, at the largest paper format, with the construction rays off — the rays explain the projection and are noise once you are transferring. Scale up from the printed cells by hand; they are not equal, so measure each one rather than stepping a fixed distance.

Can I export the distorted lattice as vector?

Yes. SVG keeps every cell edge as its own path at its own length, which matters here more than on a regular grid: the whole point is that no two cells are the same size, so anything that resamples the geometry destroys the effect.

Choosing a grid

Should I use a projector instead?

In a dark room, on a surface square to the lens, yes — it is faster. The grid method wins where a projector cannot go: daylight, a floor piece nobody can stand back from, or a wall with no throw distance. It also survives someone walking through the beam.

How is this different from ordinary one-point perspective?

One-point perspective draws a space receding from a viewer looking at a flat picture. This distorts a flat design so that it looks undistorted from ONE position — the correction is applied to the artwork rather than depicted in it. The one-point perspective generator draws what you are looking at; this draws what you paint on.

Practical matters

Does anything I load get uploaded?

No. There is no server to upload to. The photograph is read by your browser, the grid is drawn on your machine, and the export is generated locally.

Does it work for a floor piece people photograph?

That is its most common modern use. Set the viewer point where the camera stands rather than where a person walks. Set it from the intended photograph's position and the piece will read correctly in the picture and be a puzzle in the room.

Can I use it commercially for a mural commission?

Yes. The construction is centuries old and unowned, exports carry no watermark, and the terms leave the artwork entirely yours.

Open the tool Anamorphic grid generator

One position, one picture

Anamorphic perspective takes three numbers: the viewer point, the depth and the skew. Set them, then transfer. Export vector or 600 DPI. No signup, nothing uploaded.

Back to the generator →