Your pet, your kids, your house — traced into black lines ready to color. Free, unlimited, and your photo never leaves this device.
Pick a photo to start.
The photo is shrunk to at most 1,000 pixels on its longest side and turned to grayscale. For pets and people, a difference-of-Gaussians filter blurs it twice and keeps the places that are darker than their surroundings — whiskers, eyes, the outline of an ear. For houses, cars and toys, the Canny edge detector follows the strongest edges into long, clean lines. Tiny specks under a size limit are dropped, the lines are thickened to the width you pick, and the page is cropped to the drawing and centered on the sheet.
A busy background is the main reason a picture to coloring page conversion looks messy: grass and leaves become hundreds of little lines. “Remove background” runs GrabCut from OpenCV, which learns the colors inside and outside your box and paints the background white before the lines are traced.
Start with detail 3 and line thickness 1. Too many lines? Lower the detail. Missing the outline? Raise it, or switch between the Pet and Person styles. Younger kids color more happily with fewer, thicker lines.
Most “photo to coloring page” sites upload your picture to their servers to run it through an AI model — fine for a landscape, less so for a photo of your child. Here the whole process runs in your browser’s memory: there is no upload button behind the scenes and nothing is saved. Right after it opens, the page also fetches its own PDF maker (the open-source pdf-lib), so tracing, erasing, printing and the PDF work the same with the network unplugged once it has loaded. The only thing it downloads on request is the open-source OpenCV.js library (Apache-2.0), when you press Remove background.
The honest trade-off: tracing keeps the real lines of your photo, so it will not look like a hand-drawn cartoon. For a ready-made drawing, browse the free coloring pages, put a name on a page with the bubble letter generator, or finish your book with a coloring book cover.
No. The photo is opened by your own browser and turned into lines on this page; it is never sent to our server or to anyone else, and nothing is stored after you close the tab. This page also does not load session-recording tools. Once the page has finished loading you can even switch off your internet connection: tracing, the eraser, Print and Download PDF keep working. Only the background remover needs to download its free library the first time you use it.
No. It uses classic edge detection: a difference-of-Gaussians filter for pets and people, and the Canny edge detector for houses and objects. That means it traces the lines that are really in your photo — it does not redraw it as a cartoon, and it never invents details. AI generators redraw more freely, but they upload your photo and usually charge credits.
One clear subject, in focus, filling most of the picture, against a plain or blurred background, in soft daylight. Side views of pets and close-ups of faces work well. Busy backgrounds such as grass, trees or a cluttered room turn into lots of small lines: draw a box around the subject and press “Remove background” first.
You draw a box around what you want to keep and the GrabCut algorithm from the open-source OpenCV library separates it from the background by color, right in your browser. It takes a few seconds (longer on older phones). If it cuts off an ear or a tail, draw a slightly bigger box and try again, and use the eraser for leftovers.
Yes. Add up to 12 photos, give each page a title if you like, tick “Add a cover” and download one PDF with the cover and one page per photo, on US Letter or A4.
Yes — unlimited pages, no account, no credits and nothing to cancel later. There is no watermark over the drawing; each sheet only has a small footer line with the site address.
