Home › Image to Google Form

Image to Google Form: Turn a Photo or Scanned Worksheet Into a Quiz

Updated November 7, 2026 · ~8 min read

Plenty of tests still live on paper — a photocopied quiz, a scanned exam, a phone photo of last year's worksheet. Google Forms cannot read any of those. What you need is a fixed-column spreadsheet built from the printed page through OCR, then converted into Form fields — with the questions, choices, and answer key carried over exactly as printed and no AI inventing questions. This guide covers the deterministic path from an image to a Google Form.

vs. an AI quiz maker: AI quiz makers take a prompt and generate questions, often changing the answer key. Here the source is the physical page itself. OCR only reads what is printed, so a graded test stays accurate. No AI is used to create content.

Why an image is not directly importable

A photo or scan is a picture, not data. Google Forms consumes a structured spreadsheet with named columns. So the bridge is: image → OCR → fixed columns (CSV/XLSX) → Form columns → import. OCR is what turns the picture into text; the fixed columns are what Forms reliably recognizes.

The fixed-column layout Forms understand

One row is one question. This layout reshapes each parsed question into the fields Forms import.

ColumnWhat goes in it
questionThe question text as read from the page
option 1 … option 6Answer choices (leave unused blank)
correct answerThe correct option number (1–6)
pointsPoints for that question (optional)
imageOne public image URL for the question (optional)

Once the OCR output is shaped into this sheet and converted into Form columns, every row becomes a question with its original options and answer key intact.

Image to Google Form in 5 steps

1

Capture a clean image

Photograph or scan the paper test on a flat surface with good light. A sharper source means a more accurate parse.

2

Read it with OCR

The page is read into text — questions, choices, and the printed answer key — with nothing invented.

3

Arrange into fixed columns

One row per question: question text, options (option 1–6), correct answer, and points.

4

Convert to Form columns

The sheet is turned into the columns Forms reads — question, options, answer key, and points.

5

Import and grade

Load the file into the Form, then enable Settings > Quizzes with the answer key filled. Submissions grade themselves.

Diagrams travel too

Questions that depend on a figure can keep an image URL column, so the diagram moves into the Form alongside the parsed text. A geometry or graph question loses nothing in the conversion.

Frequently asked questions

Can I really turn a photo of a worksheet into a Google Form?

Yes. A photo or scan of a paper test is read through OCR into a fixed-column spreadsheet, then converted into Form fields. The questions, choices, and answer key from the printed page are carried over as parsed — nothing is invented.

Does the OCR rewrite or guess the questions?

No. OCR only reads what is printed. The conversion is deterministic: the parsed text becomes the question and options. No AI generates new questions or changes the answer key, which keeps a graded test accurate.

What kinds of sources work?

A phone photo of a paper quiz, a scanned PDF of a test, or a saved image of a worksheet all work. The cleaner the scan, the more accurate the parse — good lighting and a flat page help most.

Can image questions with diagrams be imported?

Yes. Questions that rely on a figure can keep an image URL column so the diagram travels into the Form alongside the parsed text.

How is this different from an AI quiz maker?

AI quiz makers often invent questions or change the answer key from a prompt. Here the source is the physical page itself, and only what is printed is converted. No AI is used to create content.

Put your paper tests online

Stop retyping a photocopied quiz by hand, and stop letting an AI rewrite your exam. If the questions are already on the page, send them as-is — from image to Google Form, with the printed answer key intact.

Try ExamParser →