Evening Street White Dress Portrait

A detailed prompt for generating a photorealistic portrait of a woman in a white dress sitting on a fence in an evening shopping street, with specific instructions on pose, lighting, and composition.

Prompt by @Prompt アトリエ|AI画像プロンプトView the original post ↗Published 2026-09-13CC BY 4.0

The prompt

Subject:
White one-piece dress in the evening city

Main Subject:
A vertical full-body photo of an adult woman wearing a white mini one-piece dress, sitting on a metal sidewalk fence in a shopping street at dusk. Place the person centered in the frame.

Person & Expression:
Dark brown straight hair down to the waist with thin bangs. Oval face shape, horizontal brown eyes, natural eyebrows, small nose, glossy pink lips. Twisting upper body to look back at the camera. Face turned over shoulder towards the camera. Calm expression.

Clothing & Pose:
White sleeveless fitted mini one-piece dress, grey-brown knee-high high-heeled boots with two buckles on the side, white quilted chain backpack. Sitting shallowly on the fence with both legs lowered onto the sidewalk.

Background & Light:
Night town stretching into the distance with shop signs, cars, pedestrians, and streetlights. Front flash illuminates the white clothes and legs brightly, background lights are orange bokeh. Main light source in the background is hard direct light from the front of the camera.

Composition & Camera:
3:4 vertical composition, camera is slightly low rear diagonal full-body shot. Center the person, include both boots down to the bottom, pass the fence horizontally. Focus on the person, blur the town heavily. Capture the person largely, focus on the subject, slight blur on background.

Texture & Style:
Photorealistic night snapshot. Clearly capture white cloth, quilting, grey-brown boots, and metal fence with hard flash.

Negative:
Do not cut off both knee-high buckle boots outside the frame

Generate opens this prompt in the editor, ready to run.

Model
GPT Image 2
Media
Image
Language
Japanese
Words
248
Profile / AvatarPhotographyCityscape / StreetPortrait / SelfieAbstract / Background

How to use this prompt

  1. 1Copy the prompt — fill in any placeholders first and the copy takes your version.
  2. 2Paste it into your image model, or press Generate to open it here with your own photo.
  3. 3Run it twice. Generation is not deterministic and the second pass is often the keeper.
  4. 4Finish it with the free tools: upscale, crop to the platform you need, compress before sharing.

Go beyond generating the image

Everything after the generation, on the same site.

Upscale to 4KRemove the backgroundCrop for a platformCompress before sharingAll AI apps

More from GPT Image 2

Same model, same dataset — GPT Image 2 prompts people have shared.

View all GPT Image 2 prompts →

All image categories

Use cases
Styles
Subjects