Y2K Three-Frame Selfie Collage

Create a realistic vertical collage of three close-up selfies featuring the same girl, with consistent appearance, outfit, and makeup across every frame. Use a reference image for her appearance and create a spontaneous Y2K look with direct flash

Bookmark 0

Example outputs

AI-generated image of a vertical Y2K collage of three close-up, direct-flash selfies of a girl in a vintage sweatshirt, with varied playful expressions and consistent appearance.

Prompt template

Format
Text
Inputs
Reference_image (image) Β· Required
Outputs
Three_selfie_collage (image)
Use @image1 as a strict reference for the girl's appearance. Keep her exact facial features, proportions, eye color, skin, hair, and age. Don't change her identity.

Create the most realistic vertical 3:4 collage of THREE horizontal selfies of the same size, located strictly one under the other. Use thin dividing lines and no frames. All three frames were taken by one girl on the front camera of the iPhone 17 Pro Max at close range.

IMAGE: A very voluminous vintage sweatshirt in faded dark blue with large white abstract graphics on the chest, dropped shoulders, and wide sleeves. The hair is gathered in a high, slightly disheveled ponytail, with loose strands framing the face. Match the hair color and length to the reference. Add silver ring earrings, several massive rings, and a long square milk manicure.

MAKEUP: Expressive baddie glamβ€”long, thick fox eyelashes, sharp elongated arrows, warm bronzer, bright cold pink blush, and plump lips with a brown-pink contour and wet gloss. Keep the skin realistic, with visible pores.

TOP FRAME: Very close angle, slightly from above. The girl looks straight into the lens, raising one eyebrow, with her lips folded into an ironic half-smile. One hand lightly touches her temple.

MIDDLE FRAME: The girl tilts her head to the side, sticks out just the tip of her tongue, squints her eyes, and shows a careless V sign with two fingers near her cheek.

BOTTOM FRAME: An even closer selfie, taken diagonally, with part of the crown cut off. The girl looks straight into the camera with a languid expression and slightly open glossy lips; one strand crosses her cheek.

BACKGROUND: A light wall in a home room. Hard direct iPhone flash creates sharp natural shadows behind the head. Contrasting Y2K aesthetic, slightly cold processing, digital noise, and the feeling of real, spontaneous selfies. Keep the same appearance and makeup in all frames.

NEGATIVE PROMPT: Face change, different girls, different outfits between frames, repeated emotions, extra fingers, deformed hands, plastic skin, CGI, HDR, excessive retouching, extra panels, watermarks.

β€”Ar 3:4 β€”style raw

What you'll create

This prompt creates a vertical collage of three horizontal close-up selfies of the same girl, using a reference image to preserve her appearance. It specifies a vintage sweatshirt, baddie glam makeup, varied expressions, and a spontaneous Y2K look with hard direct flash

Generation settings

Advanced settings
Negative prompt
Face change, different girls, different outfits between frames, repeated emotions, extra fingers, deformed hands, plastic skin, CGI, HDR, excessive retouching, extra panels, watermarks.

Usage instructions

Provide a reference image as @image1 so the prompt can preserve the girl's specified appearance. The three poses and expressions define the collage's variation while the outfit, makeup, and identity remain consistent; adjust those described poses if you want a different set of expressions

Frequently asked questions

What should the reference image control?

Use @image1 to guide the girl's facial features, proportions, eye color, skin, hair, and age while keeping her identity consistent

How should the three frames differ?

Use the specified raised-eyebrow half-smile for the top frame, tongue and V sign for the middle frame, and languid expression with a strand across the cheek for the bottom frame

Comments

Tried this prompt? Share what you made, compare results, or ask a question

Up to 4 results. Images up to 2 MB; MP4, WebM, or MOV videos up to 4 MB each