Editable image conversion tests and guides
Hands-on tests of screenshots, diagrams, AI artwork and document images. Each guide shows the real input, the editable result, what was measured versus reconstructed, and what to check before you rely on the export.

Canva Magic Layers alternative: take the layers out as a PSD
Canva Magic Layers splits a flat image into layers you edit inside Canva. When the layers have to leave Canva, here is what Canva documents, and two real separations exported as a layered PSD.

How to edit text in a screenshot: 3 ways compared on one headline
Cover the old words and retype, let an editor detect the text and retype in place, or redraw only the marked line. One screenshot, one word changed, four tools, measured.

How to separate a flat image into layers: Photoshop vs AI
Three ways to split one flat poster into layers — by hand in Photoshop, with an open layer-decomposition model, and with EditableImage — with the files, the timings and what each one got wrong.

Image to text with formatting: a 10-block OCR test
We tested a report image with headings, bullets, a paragraph, chart and footer. Nine text blocks kept their structure; the chart stayed a figure.

When your architecture diagram only exists as a picture
A whiteboard photo, a screenshot, a PNG in the wiki — often the only copy of a diagram is an image. What recovering it got right, and what it missed.

How to fix text in an AI-generated image: 3 ways compared
Three ways to fix a wrong word in an AI image — re-run the prompt, repaint a selection with generative fill, or mark the words and keep only those pixels.

