How to convert a photo into a cross-stitch pattern

What actually happens when a photo becomes a pattern

A cross-stitch pattern is a grid where every square is one stitch of a single floss color. Converting a photo means the image is shrunk to a grid of that size, and each cell is snapped to the nearest real thread color. That's two decisions rolled into one: resolution (how many stitches across) and palette (how many colors, and which ones).

Good converters map colors in a perceptual color space (CIE Lab) rather than raw RGB, so a "close" color actually looks close once it's stitched. They also limit the palette to real DMC floss colors instead of millions of screen colors — that's what makes the result buyable and stitchable.

Step 1 — Choose the right photo

The photo does more for the final result than any setting. Aim for:

• Sharp focus on the main subject — blur becomes muddy, indistinct stitches.
• Even, front lighting. Harsh shadows and blown-out highlights lose detail that no color count can recover.
• A subject that fills the frame. A face 200 pixels wide in a busy scene won't survive being reduced to 60 stitches.
• A simple background. A plain or blurred backdrop keeps the palette focused on the subject instead of spending colors on clutter.

Crop tightly before you convert. Cropping in-camera detail toward the subject is the single biggest quality lever.

Step 2 — Set the stitch size (width in stitches)

Pattern width is measured in stitches, not inches. The finished size depends on your fabric: finished width (inches) = stitch count ÷ fabric count. A 140-stitch-wide chart on 14-count aida finishes at 10 inches; on 18-count it's about 7.8 inches.

Rough guidance:

• 60–80 stitches — quick, gift-sized, an evening or two. Bold shapes, less fine detail.
• 100–140 stitches — the sweet spot for portraits; enough detail for a face without becoming a marathon.
• 160+ stitches — heirloom detail (fur, foliage, fabric texture), but weeks of stitching.

More stitches captures more nuance but multiplies the work: doubling the width roughly quadruples the stitch count.

Step 3 — Choose the color count (and dodge confetti)

Color count is where most auto-generated patterns go wrong. Too many colors produces confetti: dozens of single, isolated stitches scattered across the chart that are tedious to stitch and read as noise, not detail. Fewer, well-chosen colors give cleaner blocks and a piece that actually looks like your photo from a step back.

Start around 15–25 colors for a portrait. Drop lower (10–15) for solid-colored subjects or a graphic look; go higher only when the subject genuinely has smooth gradients (a landscape sky, a gray tabby's fur). See why fewer colors makes a stitchable pattern for the full reasoning.

Step 4 — Review the chart before you commit

Before exporting, zoom into the preview and sanity-check:

• Eyes and key features have a highlight and read as themselves.
• Edges separate the subject from the background cleanly.
• Color blocks are mostly contiguous, not a snowstorm of one-off stitches.

If something looks muddy, it's almost always fixable by cropping tighter, nudging the color count, or slightly increasing the stitch width — in that order.

Step 5 — Export the printable chart and floss list

A finished pattern needs three things to be stitchable: a gridded symbol chart (with bold counting lines every 10 squares and a center mark), a color key mapping each symbol to a DMC code, and a floss shopping list with how many skeins of each color to buy. StitchSnap builds a photo into exactly this — a print-ready PDF plus the DMC list — so you can start stitching from the center.