Prompts
EN

Image prompt Portraits & Fashion Customizable

Rooftop Selfie Over Shibuya Crossing in Afternoon Light

A third-party style close shot of a person holding a smartphone on a rooftop deck, with the packed scramble crossing spread below in bright crosswalks and crowds. Vivid displays and glass facades fill the background while the face and outfit stay sharp.

Model
GPT Image 2
Type
Image
Aspect ratio
1:1
Shared
17 Jun 2026
Likes on X
91
Make it yoursSwap in your own details: 2 blanks to fill

CityDevice

Image prompt

3,159 characters

Depict the person from the reference image at the moment of taking a selfie with a smartphone, on a rooftop observation deck of a high-rise commercial building from which the Shibuya scramble crossing can be seen below and in front. It is not the smartphone's exact camera view, but a selfie-style composition of that moment captured by a third-party camera at close range. Maintain the reference image's facial features, eyes, hairstyle, hair color, sense of age, the color, shape, material, and ornamentation of the outfit, and the person's atmosphere, without turning them into a different person. Do not inherit the original background, lighting, composition, or pose. Make the person the sole protagonist, in a stable 3/4-body composition from the face to around the thighs. Do not force the full body in; prioritize the fidelity of the face, upper body, outfit, smartphone, and background. The person stands on a safe rooftop observation deck, one arm extended forward holding the smartphone, gaze naturally directed at its screen. The phone appears somewhat large in the foreground but does not hide the face or outfit and is not made gigantic. The other arm is placed naturally close to the body. At the bottom edge or the left/right edges, show only a small amount of the rooftop floor and a transparent glass railing to make the shooting location clear. The rooftop portion stays within 10% of the frame. Avoid floating in mid-air, falling, and dangerous edge-standing poses. The camera is slightly in front of and above the person, in an oblique downward view over Shibuya. Do not use a fully overhead viewpoint. Place the scramble crossing large, centered and below, behind the person, clearly showing the wide crosswalks in multiple directions, the center of the intersection, and the many pedestrians crossing during the scramble. Do not make the people sparse, and do not leave the center empty. Around it, place dense commercial buildings, curved glass, store facades, and large digital displays close and large, showing at least three displays. The displays show vivid abstract imagery or human silhouettes including blue, cyan, red, magenta, and pink, without legible text, company names, or real logos. The time is afternoon with soft natural light. Roads are a natural neutral gray, crosswalks a clear white. Do not desaturate, whiten, or monochrome the background to match the outfit color; keep Shibuya's vibrancy constant through the wealth of information and color from the crowd, commercial buildings, displays, and window reflections. Avoid blown-out whites, excessive HDR, sunset, night scene, fog, and a bluish color cast. The person's face and outfit should be slightly sharper than the background. Photographic references should remain high-quality photographs; anime/illustration references should retain the original art style, lines, coloring, texture, and facial proportions. Do not convert between photo and anime. Avoid extra limbs or fingers, an unnatural smartphone, broken body connections, changing it into a generic intersection, floating in mid-air, fall depictions, legible advertising text, and real logos.

How to use it

Upload a photo of the person you want to show, since the prompt is built around keeping their face, hair and outfit intact. It was shared as a GPT Image 2 prompt. Keep the 3/4 framing, the oblique downward angle over the crossing and the glowing displays, because those carry the sense of place; change the outfit colour or the time of day if you want a different mood. The result suits a travel post or a portrait for a city-themed series.

#Portrait#Fashion#Cinematic#City & Street#Selfie

Shared publicly on X on 17 Jun 2026. All credit for this prompt and its result goes to its creator. Is this yours? Ask for an update or removal.

3,159 characters

Questions about this prompt

Which AI model was this prompt made with?

It was shared as a GPT Image 2 prompt, and the image on this page is the result the creator got with it. Other models can read the same text, but the result will look different.

Can I change the details in this prompt?

Yes. It has 2 parts you can swap: city, device. Open the prompt studio, fill them in and copy your version; anything you leave empty keeps the original wording.

Who made this prompt?

It was shared publicly on X by @ai_lifehack55 on 17 Jun 2026. The prompt and its result belong to them, and this page links to the original post.

More from @ai_lifehack55