Photo to Paint-by-Numbers
Turn a photo into numbered painting regions, a color key and a reference sheet.
Private by design
Your inputs stay on your device and are not sent to our servers. Optional usage analytics does not receive your files or tool inputs.
How to use Photo to Paint-by-Numbers
- Add a photo or try the sample. The photo keeps its original proportions.
- Choose colors, outline detail and smoothing. Review the numbered painting, coloring-style outline, color reference and key.
- Download all four sheets as PDF or save one sheet as SVG/PNG.
What are the limits of Photo to Paint-by-Numbers?
Everything is generated on your device; photos, names and designs are not uploaded. No AI model or cloud processing is used. JPG, PNG and WebP: up to 20 MB and 16 MP per photo, 50 MB combined, four photos for a book and one for other tools. Photos are normalized to a 640-pixel working copy; complex textures and low contrast may need different settings. SVG and PNG export the selected sheet; PDF includes every sheet, including separate solutions. PDF artwork is rasterized at approximately 200 dpi on A4. SVG contains vector shapes/text and embedded pixels for photo puzzles; original metadata is not retained. Smaller inputs may be needed on mobile. Cancel stops the active worker; changing controls discards stale results. Uses the same alpha compositing, grayscale smoothing and Sobel outline technique as Photo to Coloring Page. Perceptual Lab color clustering runs at a proportional 360–640-pixel working edge without upscaling. Detected photo edges are retained; only isolated one-pixel quantization noise away from edges is removed. Labels are placed inside regions with enough room to print legibly; tiny details use the color reference. Very textured inputs may need fewer colors or less detail. The color-key counts describe analysis pixels, not paint quantities. This does not identify subjects or invent missing detail.