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

How to Ask Gemini for a Trending Portrait-Style Look From Your Own Photo

Learn to write a clear style prompt covering lighting, mood, outfit and background, using only a photo of yourself.

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

Maybe you have seen portrait edits online with soft golden light, a moody backdrop, or a vintage film feel, and you wonder how to get something similar for yourself. This guide is for you if you have never asked an AI to edit a picture. By the end, you will have uploaded your own photo, written a clear style prompt, and reviewed the result.

💡 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 and the Gemini app or the Gemini website. Sign in with the account you normally use.
  • One clear photo of yourself, taken by you or with your permission from the person who took it. Avoid photos that include other people, and never upload pictures of children, because they need extra care and consent.
  • About 15 to 20 minutes. Image features may not be available on every account or plan, and some may require a paid plan. If you can't find an image option, that is usually the reason.
1
Step 1 of 5

Choose the right photo

Pick a photo where your face is clearly visible, facing the camera or close to it, with no sunglasses or heavy filters. Good lighting in the original helps, because the AI builds on what it sees. Crop out anything you don't want, such as a busy background or a stranger's arm at the edge of the frame. Save the photo to your phone or computer where you can find it easily.

Why this matters: the AI works best with a sharp, simple starting point. A blurry or crowded photo gives it more guessing to do.

✅ You'll know it worked when

you can open the photo and see your face clearly, with no one else in it.

Next step ↓
Choose the right photo
2
Step 2 of 5

Start a new chat and attach your photo

Open Gemini and start a new conversation. Look for an attachment button next to the message box, often shaped like a paperclip or a plus sign, or an option to add an image. Tap it, choose your photo from your device, and wait until a small preview of the image appears above or beside the message box. Don't send anything yet.

If none appears, your account may not have image uploads turned on yet. Try again later, or check Google's Gemini help pages for current availability.

✅ You'll know it worked when

a thumbnail of your photo sits in the message box, ready to send with your next message.

🔄 If it looks different

If you don't see an attachment button, look for an option labeled as adding a photo or image in the menu near the message box.

Next step ↓
Start a new chat and attach your photo
3
Step 3 of 5

Write a clear style prompt

This is the most important step. A prompt is simply the written instruction you give the AI. Describe four things: the lighting, the mood, the outfit, and the background. Be specific but keep it to a few sentences.

💬 Try typing thisType your own details into the dashed boxes, then copy."Turn this photo into a portrait with , a feeling, , and a . Keep my face and hairstyle as close to the original as possible."➤

Replace the bracketed parts with your own choices. Why does this work? Lighting tells the AI where the light comes from, mood sets the feel, outfit describes clothes, and background sets the setting. The last sentence asks the AI to keep your face recognizable.

Send the message, then wait a little while. Image creation can take some time, and the result may appear as a new image below your message.

✅ You'll know it worked when

Gemini returns a new image based on your photo, and the lighting and background look roughly like what you asked for.

Next step ↓
GeminiTurn this photo intoa portrait with[lighting style…
✏️ Sketch — your screen may look a little different
4
Step 4 of 5

Check the result against your original

Look at the new image next to your original. Check three things: does it still look like you, did the requested lighting, mood, outfit and background appear, and did anything change that you didn't ask for, such as your eye color, teeth, or a tattoo. AI image tools often make small changes, and some results will look unnatural in the hands, hair edges or background.

If the result misses the mark, don't start over from scratch. Reply in the same chat with a short follow-up, such as: "Keep the same pose and face, but make the lighting warmer and remove the extra jewelry." Small, specific changes usually work better than rewriting the whole prompt.

💬 Try typing this"Keep the same face and pose, but make the lighting warmer and the background a little less blurred."➤
✅ You'll know it worked when

your follow-up produces a version closer to what you pictured, and the face still looks like yours.

Next step ↓
GeminiKeep the same faceand pose, but makethe lighting warmer…
✏️ Sketch — your screen may look a little different
5
Step 5 of 5

Save the image and keep your edits honest

When you like a result, save it. Look for a download option on the image itself, often a down-arrow icon or a menu with a "save" or "download" choice. Hold the image and choose to save it to your photo library if you're on a phone. On a computer, right-click the image and choose to save it, or use the download option if you see one.

Before you post it anywhere, consider labeling it as AI-edited, especially if people might think it's an unedited photo. Don't use an AI portrait for identity documents, profiles where people need to verify you, or anything that could mislead others about who you are.

💬 Try typing this"AI-edited portrait, created from my own photo."➤
✅ You'll know it worked when

the image is saved to your device and you can open it outside the chat.

Save the image and keep your edits honest
⚠️ Common mistakes
  • Uploading other people's photos. Using a picture of a friend, family member or child without permission is unfair to them. The fix: use only your own photo, and if you ever need a picture of someone else, ask them first and let them decide.
  • Writing a vague prompt like "make me look cool." The AI has to guess what you mean, and you may get something you don't like. The fix: name the lighting, mood, outfit and background in one sentence each.
  • Trusting the first result as final. The first image often has small errors. The fix: compare it to your original, then send one or two targeted follow-up requests in the same chat.

One honest limit: Gemini can change the look of your photo, but it cannot guarantee a perfect likeness of you, and it cannot know your personal taste without you describing it.

🚀 Try it now

Pick one photo of yourself from your phone, open Gemini, attach it, and send this message: "Turn this photo into a portrait with soft window light and a calm, confident mood. Keep my face and hairstyle as close to the original as possible." Then look at the result next to your original and decide what to change.

🔧 If you get stuck

Why can't I edit images in Gemini?

Why: Editing images isn't available to users under 18. To use this feature, you must be signed in to Gemini Apps.

What to do: To generate and edit images, you must be 18 or over. To use this feature with a personal account, you must meet the age requirements and be signed in to Gemini Apps.

Source: Generate & edit images with Gemini Apps - Computer

Gemini won't combine my photo with another picture.

Why: Nano Banana 2 Lite isn't designed for multiple image references or for multiple edits. Nano Banana 2 can accept multiple reference images.

What to do: Switch your Gemini model before you create your image request. Nano Banana 2 is selected when you make an image request with the Gemini model set to Flash or Pro.

Source: Generate & edit images with Gemini Apps - Computer

How do I edit a photo I already have, not one Gemini made?

Why: You can upload an image from your device that you want to edit. You can upload an image to edit or ask Gemini to edit a generated image.

What to do: Upload an image and ask Gemini to make edits. In the text box at the bottom, enter your prompt, then click Submit.

Source: Generate & edit images with Gemini Apps - Computer

I can't redo my image with Pro anymore.

Why: If you reach your daily quota of Nano Banana 2 images, you can't redo any additional images with Nano Banana Pro. Nano Banana Pro is available with a Google AI Plan, and paid subscribers can use it to regenerate an image.

What to do: With a Google AI Plan, create an image with Nano Banana 2 or Nano Banana 2 Lite. At the bottom right, click More, then Redo with Pro.

Source: Generate & edit images with Gemini Apps - Computer

🎯 Next guideHow to Use Gemini to Make Your LinkedIn Photo Look Professional at HomeLearn to start with a clear, well-lit photo and ask Gemini for a gentle edit that still looks like you.

❓ 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 and the Gemini app or the Gemini website. Sign in with the account you normally use.
  • One clear photo of yourself, taken by you or with your permission from the person who took it. Avoid photos that include other people, and never upload pictures of children, because they need extra care and consent.
  • About 15 to 20 minutes. Image features may not be available on every account or plan, and some may require a paid plan. If you can't find an image option, that is usually the reason.
What mistakes should I avoid?
  • Uploading other people's photos. Using a picture of a friend, family member or child without permission is unfair to them. The fix: use only your own photo, and if you ever need a picture of someone else, ask them first and let them decide.
  • Writing a vague prompt like "make me look cool." The AI has to guess what you mean, and you may get something you don't like. The fix: name the lighting, mood, outfit and background in one sentence each.
  • Trusting the first result as final. The first image often has small errors. The fix: compare it to your original, then send one or two targeted follow-up requests in the same chat.

One honest limit: Gemini can change the look of your photo, but it cannot guarantee a perfect likeness of you, and it cannot know your personal taste without you describing it.

Keep reading

🔥 See all of this week’s AI trends →

📬 The week’s AI, in your inbox

One friendly email every Sunday — the 5 stories that mattered, in plain English. No spam, unsubscribe anytime.

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.

📘 Prefer it in one file? Photos and Images: The AI Pack (30 guides, free printable PDF)

← Back to all stories