Paper Animator

Turn one photo into a torn-paper collage animation.

All processing happens in your browser. Your files are never uploaded to any server.

Stage

🖼️

Drop an image here

or click to choose  ·  PNG · JPEG · WebP · GIF (first frame)

Waiting for an image.

Export

Export formats are exactly what the browser can encode. PNG is rasterised from the canvas. WebM is encoded frame by frame from the canvas via MediaRecorder, and takes about as long as the clip itself. WebM is the only animation container this page can produce — MP4 and GIF are not offered because they are not exported here.

Controls

10
2.0
0.55
0.50
0.40
0.45
Flood-fill the background away from the image borders. Works when the subject sits on a reasonably even background.
Backing sheet colour behind the torn edge
0.50
3.0s
What this page can and cannot do
Can:
  • Read a single still image in any format your browser can decode.
  • Synthesise a torn / deckle / shredded paper edge in real time.
  • Composite a paper sheet, a drop shadow, loose fibres and a raw paper core along the tear.
  • Flood-fill an even background away from the image borders (Cutout), with a tolerance slider. The panel tells you what percentage it actually removed.
  • Export the still frame as PNG, and the animation as WebM.
Cannot — and is therefore not offered in the UI:
  • Video input. A video file cannot be read here; only the first frame of an animated GIF is used.
  • MP4 / MOV / GIF export. The browser ships no free encoder for those; only WebM is produced.
  • Sound or music. There is no audio pipeline on this page.
  • Object recognition. Cutout is a colour flood-fill from the borders, not AI segmentation — a busy, textured or gradient background will not separate cleanly.
  • Upscaling, enhancement, face detection or any AI model. No model is loaded or called.

Paper animator — what it supports, how it counts, and common questions

Supported input and output

This paper animator works on one still image at a time. Drop in a PNG, JPEG, WebP or GIF — an animated GIF is read from its first frame. The image is decoded in this tab and never leaves your device, so there is no upload, no render queue and no account to create.

Everything on the right is live. Edge thickness, edge softness, tear intensity, detail, fibre grain, shadow, cut style, cutout tolerance, paper tone, motion strength, duration and frame rate all re-render the paper animation the moment you move them, so you can judge a torn-paper look by eye instead of by description.

When the paper cut-out looks right, download the still frame as a PNG, or download the animation as WebM. The PNG is rasterised straight from the canvas; the WebM is encoded frame by frame in your browser, so the export takes about as long as the clip itself.

How the numbers are counted

Nothing here is estimated by hand. The page measures its own work and prints the result under the canvas.

  • Working size. The image is fitted to a working canvas of up to 1600 px on the long side, and very small images are scaled up so the tear has room to exist. The status line reads back both sizes, for example Loaded 3840×2400 → working at 1600×1000.
  • Cutout. With cutout on, the page counts the pixels it actually flood-filled away and reports the share of the whole frame, for example Cutout removed 66.3% of the frame as background. If that number is near zero, or near the whole frame, the line says so and tells you which way to move the tolerance slider.
  • Frames. Duration (1–10 seconds) multiplied by the chosen frame rate (24, 30 or 60 fps) is the frame count for the animation export.
  • Export. After a download the page reports the real file size, and for WebM also the frame count, the frame rate it actually reached and the real time it took — for example 30 frames @ 10 fps, recorded in 3.3s. If this machine cannot hold your chosen frame rate, the status line says so rather than pretending.

Frequently asked questions

Is Paper Animator free, and do I need an account? It is free, and there is no account, no email and no sign-up. Nothing is metered, limited or queued.

Where is my image processed? In your browser. The image is decoded in this tab, the whole torn-paper pass runs on your own machine, and the file is never uploaded to any server.

What image formats can I use? PNG, JPEG, WebP and GIF. An animated GIF is read from its first frame.

What can I download? A still frame as PNG, and the animation as WebM.

Is there a watermark or an export limit? No watermark and no limit — exports are unmarked and nothing is counted.

Does every image get the same tear? No. The torn edge is synthesised from your image rather than stamped from a fixed template, and Shuffle tear pattern gives you a fresh tear whenever you want one.

Can it cut out the background? Yes, with the Cutout toggle. It flood-fills an even background inward from the image borders and the status line reports the exact percentage it removed. It is a colour flood-fill, not AI segmentation, so it needs the subject to sit on a reasonably even background.

Does it work on a phone? Yes. You can choose or drag an image, drag the sliders and tap the buttons on a phone-sized screen.