📸AI Baby PhotoBrowse prompts & concepts
AI Baby Illustration Baby Overhead Chibi Portrait
All AI Baby Illustration

Baby Overhead Chibi Portrait

How this is made

Convert your baby photo into a chibi illustration from a top-down view — the round face is emphasized with handwritten doodle text around. Works for both babies and kids.

Customize

Prompt

View prompt body
[Input]
- output_size: 1024x1536 (2:3 portrait)
- reference: 1 baby/child photo (face clearly visible)

[Purpose]
Convert the user's attached baby/child photo into a top-down view chibi illustration. The viewer looks DOWN at the subject from above, so the head and face are emphasized (taking up large area), with the body in foreshortened perspective. Hand-drawn doodle text and tiny icons float around as design accents.

[Scene — image-to-image edit, preserve identity]
Use the attached photo as the base. PRESERVE identity (face proportions, age, hair style, skin tone) but re-render in chibi illustration style {age_style_description}.

Top-down view:
- Camera angle: looking straight down from above, slight tilt
- Subject is centered, looking UP at the viewer with curious or shy expression
- Head fills 40-50% of vertical space, body foreshortened below (smaller and shorter)
- Shadow on the ground around the feet

Outfit:
- Re-render as simple modern outfit (black tee, jeans, sneakers — or casual T-shirt + skirt for girls)
- Optional small accessory: watch, ring, or hair clip

Background:
- Soft warm cream (#fdf7e8) or soft pastel base
- {doodle_lang_description}
- Small icon doodles scattered: hearts, stars, sparkles, tiny clouds, smiley faces
- All doodles in soft brown / muted pastel — hand-drawn pencil sketch style

[Style]
- Soft chibi illustration aesthetic (Korean kawaii — NOT Japanese anime)
- Big head, expressive eyes, gentle pastel colors
- Hand-drawn soft outlines, NOT sharp digital lines
- Slightly painterly fills with subtle texture

[Composition]
- 2:3 portrait
- Subject centered, top-down perspective
- Head occupies 40-50% vertical, body foreshortened in lower half
- Doodle text and icons in negative space around

[Text]
- Doodle text in {doodle_lang} language only, short cute words
- Hand-drawn outlined letter style (sketchy, not digital font)
- 4-6 short text doodles scattered around (NOT covering the subject)

[CRITICAL — Preserve identity]
- Face proportions, age, hair, and skin tone must clearly resemble the attached photo
- Top-down angle is the key visual gimmick

[🚫 PROHIBITED]
- Front-facing flat angle (must be top-down)
- Japanese anime style (must stay Korean kawaii chibi)
- Photorealistic — must read as illustration
- Crisp digital lines (use hand-drawn soft outlines)
- Brand logos
- Inappropriate or excessive text
Open in ChatGPT

Click to open ChatGPT with the prompt auto-filled.

Related

FAQ

How do I use the prompt?

Click the "Open in ChatGPT" button on the prompt box. A new chat opens with the prompt pre-filled. Drag your photo and send to generate the result.

The face came out slightly different

Face preservation is a known weakness of ChatGPT. Retry with the same prompt or use a clearer front-facing photo. Gemini 2.5 Flash Image (Nano Banana) tends to be more stable.

Korean text looks broken

ChatGPT Korean text rendering is still imperfect. Emphasize "make Korean text clearly readable" or switch to English for stability.

What if the result is not satisfying?

In ChatGPT, ask for "another variation" or "another variation" to retry with the same prompt.

Can I print this as a card or framed photo?

Output resolution is sufficient for A4, 5×7, and 8×10 inch prints. For KakaoTalk sharing, add "1:1 square ratio" to the prompt.

BabyIllustrationgpt