Add Text to an Image: How PhotoGPT Places Text Behind Any Subject

You have a photo. The photo is fine. The photo is missing something, but you don't know what. Maybe a quote. Maybe a price tag. Maybe the name of the person in the shot. You open Canva. You click "Add Text." The text sits on top of the photo like a sticker. Done. Functional. Forgettable.

Every text-on-photo tool works that way. The text is always on top. You can move it around, change the font, add a drop shadow. But it is always on top. The eight top results for "add text to image" do the same thing, with slightly different sticker shapes.

PhotoGPT does the same thing, plus one option. The text can go behind the subject. The AI removes the background of the main subject when you upload the photo. After that, the text can sit in 3D space behind the person, the product, the pet. The subject's silhouette occludes the text where they cross it. You get text that emerges from behind the subject, not a sticker pasted on top of it.

The five text designs below show what to do with it.

What "Add Text to Image" Actually Means in PhotoGPT

PhotoGPT's text-in-image tool is a browser-based text editor, not an image generator. You upload a photo, the AI detects and removes the background of the main subject, and a canvas appears with the subject on transparent. You then place one or more text boxes on the canvas, with full control over:

• Font family (Playfair Display, Inter, Bebas Neue, and 30+ others, all loaded in-browser)
• Font size, weight, color, opacity
• Letter spacing
• Shadow color and size
• Rotation and 3D tilt (X and Y axis independently)
• Position (drag anywhere on canvas)
• Placement: either "behind" the subject (text occluded by the subject's silhouette) or "overlay" on top of it

When you save, the canvas composites to a PNG. The text is baked into the image at the rendered position. No watermark. No "made with PhotoGPT" footer.

The Unique Part: Text Behind a Subject

The other seven text-on-image tools in the SERP above put text on top of the photo. That is the entire product. PhotoGPT adds one option: text behind the subject. When you place text behind a person, product, or pet, the subject's silhouette occludes the text where they cross it. The text appears to physically sit in 3D space behind the subject.

This works because the AI removes the subject's background automatically when you upload. You do not need to do the masking. The transparent background is what allows the layer order to work.

This is what the visual is showing. A person standing. Text "EVERY STEP COUNTS" behind them. The text crosses behind their torso. The text emerges around the edges of their body, with their silhouette clearly occluding it where they stand.

Three places this is useful in practice:

How to Add Text to an Image (Step-by-Step)

The full workflow takes about 2 minutes once you have a source image, longer the first time while you explore the controls.

Step 1: Upload Your Photo

Navigate to PhotoGPT's text-in-image tool. Drop any PNG, JPG, or JPEG up to 10MB. The tool supports the main formats you would have on your phone or camera. The first time you upload, the AI analyzes the image to detect the main subject. This takes 5 to 15 seconds. You will see a "Analyzing..." overlay with a particle effect while it processes.

Step 2: Inspect the Background Removal

Once analysis finishes, the subject is on transparent. The subject is the same image, but the background is gone. The text editor has placed the subject on a transparent canvas. You will see only the subject silhouette in the canvas preview.

If the AI removed the wrong thing (a person in the background instead of the foreground subject), the tool cannot recover. Upload a different image where the subject is more clearly dominant. PhotoGPT's AI subject detection is strong but not perfect for busy scenes.

Step 3: Add Your First Text Box

Click "Add Text" in the desktop sidebar. A text box appears with the word "Edit" as placeholder. Double-click to enter editing mode and type your text. The default font is Playfair Display at 300px. The default color is a warm off-white. The default position is roughly center.

Step 4: Choose Placement (The Key Decision)

In the sidebar, find the "Placement" control. There are two options:

• Behind subject: text is occluded by the subject's silhouette. Use this when the text should appear to be in 3D space behind the person.
• Overlay: text sits on top of everything. Use this for captions, labels, callouts.

For portrait quote images, behind subject is the visual that makes the result feel like a designed object, not a template.

Step 5: Adjust Style, Position, Rotation

The sidebar exposes all the text controls:

• Font: scroll through the dropdown. Playfair Display for elegance, Bebas Neue for impact, Inter for modern minimal, Permanent Marker for handwritten feel.
• Color: pick a color that contrasts the subject's silhouette area. White on dark backgrounds, dark on light backgrounds, brand color for campaigns.
• Rotation: drag the rotation handle to angle the text. A 5 to 10 degree tilt reads as intentional. A 0 degree tilt reads as flat.
• Tilt X/Y: gives 3D perspective. Subtle is better. 5 to 15 degrees is enough to add depth without looking gimmicky.
• Shadow: adds a soft drop shadow that helps text pop against busy backgrounds. Default settings work for most cases.

Drag the text box to position it where you want it. The bounding box has handles on each corner for resizing and a handle above the box for rotation. Pinch-to-zoom works on touch devices.

Step 6: Add a Second Text Box (Optional)

For layered designs, add a second text box. For example: a large quote behind the subject ("STAY HUNGRY"), and a smaller attribution on top ("Steve Jobs, 2005"). Multiple text boxes each have their own style and placement.

Step 7: Save

Click "Download." The file saves as a PNG at the original image's resolution, no watermark. The output is ready to post on any platform.

5 Text Designs You Can Apply to Your Own Photos

These are not "prompts." The text-in-image tool does not generate anything. These are five starting points (text + style + placement) you can drop into the editor with your own image. Adapt the text, change the font, adjust the position. The recipe is the structure, not the literal words.

1. Portrait Quote, Behind Subject (Centered Torso)

Text: "STAY WHERE YOUR ENERGY IS"
Font: Playfair Display, weight 700, size ~280px
Color: #ffffff (white)
Shadow: #000000, size 30
Rotation: 0°
Tilt: TiltX 0, TiltY 0
Placement: behind subject, positioned at 50% horizontal, 30% vertical

Use for: Instagram quote posts, LinkedIn carousel frames, personal-brand hero images. The text frames the subject without competing with the face.

2. Lower-Third Overlay (Subtitle-Style)

Text: "Live in the moment"
Font: Inter, weight 500, size ~120px
Color: #ffffff
Shadow: rgba(0,0,0,0.6), size 20
Rotation: 0°
Placement: overlay, positioned at 50% horizontal, 12% vertical (lower third)

Use for: social media caption overlays, behind-the-scenes photos where you want a subtitle, photojournalism-style. The text is the caption, not the design element.

3. Tilted 3D Perspective (Editorial / Fashion Feel)

Text: "STILL LIFE"
Font: Bebas Neue, weight 800, size ~320px
Color: #fffaaa (warm cream)
Shadow: rgba(180,0,0,0.4), size 50
Rotation: 0°
Tilt: TiltX 8, TiltY 6
Placement: behind subject, positioned at 30% horizontal, 50% vertical

Use for: editorial / fashion / album-cover feel. The 3D tilt adds perspective. The warm cream text on a darker background reads as premium.

4. Subtle Lower-Third Quote (Less Aggressive)

Text: "In the moment"
Font: Permanent Marker, weight 400, size ~140px
Color: #ffffff
Shadow: rgba(0,0,0,0.3), size 12
Rotation: -2°
Tilt: TiltX 0, TiltY 0
Placement: overlay, positioned at 50% horizontal, 15% vertical

Use for: casual, personal, lifestyle. Permanent Marker reads handwritten, not "designed." The -2° rotation gives it movement without being a "fashion" tilt.

5. Multi-Box Layered (Quote + Attribution)

Text Box 1 (behind subject): "FOLLOW THE WORK"
Font: Playfair Display, weight 700, size ~260px
Color: #ffffff
Placement: behind subject, 50% horizontal, 35% vertical
Tilt: 0°

Text Box 2 (overlay, smaller, lower-right): "personal note, 2024"
Font: Inter, weight 400, size ~60px
Color: rgba(255,255,255,0.7) (semi-transparent white)
Placement: overlay, 75% horizontal, 80% vertical

Use for: layered designs where the quote is the hero and a small attribution sits below. Common in album covers, yearbook-style photos, or branded posts.

Each recipe above works as a starting point. Open the editor, paste the text, choose the placement, adjust to taste. The structure is what matters; the literal values are starting points.

When PhotoGPT's Text-in-Image Wins vs the Other Tools

The right answer is to know what job you are doing before you pick the tool. Use PhotoGPT when text-behind-subject is the move. Use Canva, Picsart, or Adobe Express when the standard top-of-image text workflow is enough.

For a searcher shopping "add text to image" with the intent of "I want text to look like it's part of the image, not pasted on top," PhotoGPT is the only mainstream tool that does this. For the searcher with a basic top-of-image text need, the 7 other tools are faster.

Workflow Pairings (After You Finish Your Edit)

Once you have your edited image, the rest of the PhotoGPT workflow handles the rest:

• Run it through the photo editor for any color matching or final contrast adjustment.
• Upscale it if you want hi-DPI for print or for an Instagram profile image (2x on the original export hits print-ready for most uses).
• Cross-link your edits in Flow if you are building a content drip campaign (e.g., 30 days of quote posts).

For more on the post-production workflow, see our AI image upscaler guide, which covers the same editor + upscale pipeline for any AI-generated image.

FAQ

How do I add text to an image without Photoshop?

Use a browser-based text editor like PhotoGPT's text-in-image tool, Canva, Picsart, or Adobe Express. Upload the photo, position the text, save as PNG. No software to install, no subscription required for the basic flow.

What is the best text-on-photo editor?

For standard top-of-image text, the major options are Canva, Picsart, PicMonkey, Phonto, and Adobe Express. For text-behind-subject (the text appears integrated with the person's silhouette), PhotoGPT's text-in-image tool is the only mainstream option that does this with automatic background removal.

Can I add text to a photo on my phone?

Yes. The PhotoGPT text-in-image tool runs in the browser, including mobile browsers. The editor has touch-friendly handles for dragging, resizing, and rotating text. The output saves to your phone's downloads folder.

Does the text-on-photo tool cost anything?

PhotoGPT includes a free tier with limited credits. Text-in-image is one of the included features. Pricing details for higher tiers are here.

Will the tool keep my uploaded image private?

Yes. Uploaded images are processed in your browser session and are not stored on PhotoGPT servers after the session ends. The output PNG lives in your downloads, not in any gallery.

Can I add multiple text boxes to one image?

Yes. Each text box has its own font, color, placement (behind subject or overlay), and rotation. A common pattern is one large quote behind the subject plus a small attribution on top. Each text box is independently edited.

Try It on Your Next Post

The fastest way to know whether this workflow fits you: upload one of your own photos, place one piece of text behind the subject, save. The first attempt takes about 5 minutes, the second about 2.

Open the Text-in-Image Editor