Step-by-step
📸 Gemini
Google
📘 Step-by-step 📘 Google · Gemini↗ Gemini · Official sitebeginner 🏠 Everyday life

How to Turn a Selfie Into a Polaroid-Style Portrait With a Gemini Prompt

A four-part prompt template (subject, outfit, lighting, frame) so your Gemini-generated image looks like a vintage Polaroid — and still looks like you.

6 min read·October 17, 2026

This guide was written by AI. It passed automated fact and quality checks, and our editor spot-checks published guides every day.

See AI World HQ first on GoogleAdd us as a preferred source and our stories show up higher in Google’s Top Stories.Add on Google

You have a selfie you like — the angle works, the light was kind, you actually felt like yourself. Now you want it to look like it was shot on a Polaroid camera in 1997: soft flash, faded colors, a chunky white border, maybe a scribbled caption. You can describe that look in words, and Google's Gemini can turn those words into an image. The trick is giving Gemini enough structure that it has something specific to work with — not just "make me a polaroid."

💡 Tip: tap a step’s number when you finish it — a green tick appears and your browser remembers how far you got.

✅ Before you start
  • A Google account you can sign in with.
  • Access to Gemini — either the website (gemini.google.com) or the Gemini mobile app. Image creation may sit behind a paid Google AI plan on some accounts, so check your plan if the image option is missing.
  • One clear, well-lit selfie saved on your device (phone or computer). Front-facing, no sunglasses, plain background works best.
  • About five minutes total, plus a little patience — getting the look you want usually takes two or three rounds.
1
Step 1 of 5

Open Gemini and start a fresh chat

Open gemini.google.com in a browser, or launch the Gemini app on your phone. Sign in with the same Google account you use for Gmail or YouTube if asked. Once you're in, look for the "New chat" button — usually a plus icon, a pencil, or the words "New chat" near the top of the chat list. Tap or click it.

You should see a clean, empty chat box waiting for your first message. Starting fresh matters: image generation works best when the prompt is the only instruction Gemini has to focus on, with no leftover instructions from a previous conversation.

💬 Exampledon't type anything yet — just confirm the chat is empty.

Either way, you should land on a chat screen with a text box at the bottom.

🔄 If it looks different

on older versions or some Android setups, the entry point is "Gemini" inside the Google app (the "G" logo) rather than a separate Gemini app.

✅ You'll know it worked when

**** you see an empty message field and no previous conversation threads.

Next step ↓
Geminigemini.google.…
✏️ Sketch — your screen may look a little different
2
Step 2 of 5

Attach your selfie as a visual reference

In the empty chat box, look for an attachment icon — usually a paperclip, a plus sign, or an image icon near the text field. Tap it and pick "Upload from device" (or a similar option). Choose the selfie you prepared. Wait for the image to finish uploading — you'll see a small thumbnail of your photo appear in the message box.

Attaching a reference photo gives Gemini something to anchor the look to: your general face shape, hair, skin tone, and what you were wearing. Without a photo, Gemini is guessing what you look like from words alone, and the result rarely feels like you.

💬 Exampledon't type a prompt yet — just attach the selfie and watch the thumbnail appear.

If you can't find an upload button at all, your current Gemini surface may not support image inputs — in that case, skip this step and rely on a very detailed text description in Step 3.

🔄 If it looks different

on some Gemini surfaces, the attachment icon is hidden inside a "+" or "Tools" menu next to the text box.

✅ You'll know it worked when

**** a small image preview of your selfie is sitting in the chat box, ready to be sent with a prompt.

Next step ↓
GeminiUpload from de…
✏️ Sketch — your screen may look a little different
3
Step 3 of 5

Write the polaroid prompt using the four-part structure

Click into the text box and type your prompt using this structure: (1) subject, (2) outfit, (3) lighting, (4) frame. Think of these four slots as a recipe — each one nudges the image in a specific direction. Skip one and the result drifts toward generic.

Here's the template, with each slot filled in as an example:

💬 Example*"Polaroid photo of a young woman with shoulder-length wavy hair and a small nose ring, soft half-smile. She is wearing a cream knit sweater and silver hoop earrings. Direct on-camera flash against a dark indoor background, slight red-eye, no fill light. White Polaroid 600 border with a handwritten caption 'spring break '98' in blue pen across the bottom. Soft focus, slightly overexposed, faded 90s film colors, square 1:1."*

You can swap the gender, hair, outfit, lighting, or caption — just keep all four slots filled. The frame slot is the most important for the Polaroid look: without "Polaroid," "white border," "handwritten caption," and "faded film colors," Gemini produces a normal photo, not a vintage one. Press Enter (or the send arrow) to send.

If your prompt returns only text and no image, look in the input area for that option and try again with it selected.

The image should have a white border, look slightly washed-out, and roughly match the four things you described.

🔄 If it looks different

on some Gemini surfaces, image generation lives behind a separate "Create image" or "Image" toggle instead of happening automatically.

✅ You'll know it worked when

**** Gemini generates an image (not just text) within a few seconds.

Next step ↓
GeminiEnter
✏️ Sketch — your screen may look a little different
4
Step 4 of 5

Review the result and decide what to fix

Look at the generated image and compare it to your selfie. Don't judge it yet — just notice three things: (a) Does it look like you, even vaguely? (b) Does it actually look like a Polaroid? (c) Is the outfit, lighting, or pose way off?

Most first attempts miss on at least one of these. That's normal. Gemini generates one strong guess per prompt — it doesn't "see" your feedback the way a designer would. The way to steer it is to rewrite your prompt with a small change and send it again as a fresh message.

If the face is wrong but the style is right: add more face details (eye color, jaw shape, hairline, distinguishing features like glasses or freckles). If the style is wrong but the face is close: keep the face description and rewrite the lighting and frame slots with stronger Polaroid cues — try "direct on-camera flash," "warm color cast," "thick uneven white border."

💬 Example*"The face looks a bit too round — change to: 'oval face, high cheekbones, straight nose, dark brown eyebrows.' Also add: 'visible film grain, slightly crooked white border like a real Polaroid.'"*

Two or three rounds is usually enough.

✅ You'll know it worked when

**** the next image lands closer to what you imagined.

Next step ↓
GeminiThe face looks a bittoo round — changeto: 'oval face…
✏️ Sketch — your screen may look a little different
5
Step 5 of 5

Save the image to your device

Hover over or long-press the generated image. Look for a download button (usually a downward arrow icon), or a "Save" or "Download" option in a small menu that appears. The image saves to your phone's Photos or Gallery folder, or your computer's Downloads folder.

💬 Exampledon't type anything — just click download and check your Photos or Downloads folder.
🔄 If it looks different

on the mobile app, you may need to tap the image first to open it full-screen, then look for a share icon (a square with an arrow pointing up) and pick "Save image." If you don't see any download option, a screenshot cropped to the image works in a pinch.

✅ You'll know it worked when

**** the Polaroid portrait shows up in your phone's gallery or computer's Downloads, ready to send.

Save the image to your device
⚠️ Common mistakes
  • Skipping the frame details. A prompt that says "make this a polaroid" without mentioning "white border," "handwritten caption," or "faded film colors" usually produces a normal photo with a filter. The frame slot is what sells the look — be specific.
  • Describing yourself in a way Gemini can't anchor to. Saying "make it look like me" without attaching a selfie or describing concrete features (hair color, face shape, glasses, freckles) gives Gemini nothing to work with. Either upload a reference photo or write at least three concrete face details.
  • Re-sending the exact same prompt. Image generation is one-shot per prompt — sending the same prompt twice gives you two slightly different guesses, but doesn't "fix" the first one. Change one slot (usually lighting or face details) each round.
🚀 Try it now

Open Gemini and paste this exact prompt, then swap the bracketed parts for your own details:

"Polaroid photo of [describe yourself in two sentences — face, hair, expression]. Wearing [one clothing item]. Direct flash, dark background. White Polaroid border with caption 'first try' in blue pen. Faded 90s film colors, soft focus, square."

🎯 Next guideHow to Generate Images with Google Gemini: A Beginner's GuideCreate beautiful, custom visuals for work or play in minutes using Google's free conversational AI.

❓ Quick questions

How long does this take?

About 6 minutes — the guide has 5 steps, and you can tick each one off as you go.

Which tool do I need?

This guide uses Google Gemini — but the approach works very similarly in other AI assistants.

Do I need to prepare anything?
  • A Google account you can sign in with.
  • Access to Gemini — either the website (gemini.google.com) or the Gemini mobile app. Image creation may sit behind a paid Google AI plan on some accounts, so check your plan if the image option is missing.
  • One clear, well-lit selfie saved on your device (phone or computer). Front-facing, no sunglasses, plain background works best.
  • About five minutes total, plus a little patience — getting the look you want usually takes two or three rounds.
What mistakes should I avoid?
  • Skipping the frame details. A prompt that says "make this a polaroid" without mentioning "white border," "handwritten caption," or "faded film colors" usually produces a normal photo with a filter. The frame slot is what sells the look — be specific.
  • Describing yourself in a way Gemini can't anchor to. Saying "make it look like me" without attaching a selfie or describing concrete features (hair color, face shape, glasses, freckles) gives Gemini nothing to work with. Either upload a reference photo or write at least three concrete face details.
  • Re-sending the exact same prompt. Image generation is one-shot per prompt — sending the same prompt twice gives you two slightly different guesses, but doesn't "fix" the first one. Change one slot (usually lighting or face details) each round.

Keep reading

📘 Turn a casual selfie into a clean LinkedIn headshot with one Gemini prompt A copy-paste prompt template that swaps a phone selfie for a tidy, professional-looking headshot with a simple background. 📰 How to Write AI Prompts That Look Like Real Polaroid Photos A practical, step-by-step guide to getting film-like results from any text-to-image AI 📘 How to Generate Images with Google Gemini: A Beginner's Guide Create beautiful, custom visuals for work or play in minutes using Google's free conversational AI. 📘 How to Build a Simple Weekly Schedule in Google Sheets with Gemini Describe your week and get a ready-to-use template with time blocks, color categories, and hours-totaling formulas 📘 Using Gemini in Gmail and Docs to save time Learn how to let Google’s Gemini AI quickly summarize email threads, draft replies, and polish documents – all from the tools you already use. 📘 Turn a messy review into a polished testimonial with Gemini Quickly transform rough customer feedback into compelling, professional testimonials for your website or marketing materials using Google Gemini.
Was this helpful?

✦ Generated by AI in AI World HQ's automated newsroom, written in plain language. Checked by automated fact and quality gates, and our editor spot-checks published guides every day. Spot a mistake? Use the buttons above.

← Back to all stories