mmindwiki
sitespromptsgithubaboutlog insubmit a find
mindwiki/#image-generation

#image-generation

2 items tagged "image-generation".

prompt

create a subway action scene

Create a 30-second, 1080p ultra-realistic Korean subway action-comedy scene using the uploaded images as exact visual references; keep characters, faces, hair, outfits and environment consistent. Young Korean woman, early 20s, long straight black hair, green cropped long-sleeve top, loose blue jeans, silver jewelry; confident and expressive. Young Korean man, short black hair, gray button-up shirt, dark trousers and white sneakers; playful, then shocked and embarrassed. She rushes down the subway stairs toward a waiting train, but the doors close just as she reaches them. The man inside sees her through the glass and laughs at her; she stares at him angrily and says, “Are you seriously laughing?” He laughs again, so she suddenly punches the glass beside the door, cracking and breaking the safety glass realistically with no injury or gore. She reaches through the broken opening, grabs his shirt and pulls him safely out onto the platform; he lands awkwardly on the floor in shock. She calmly climbs through the same broken opening into the train, straightens her clothes and walks inside as stunned passengers watch. The man remains sitting outside, looking confused and humiliated while passengers stare at him through the windows. One passenger starts laughing, then everyone bursts into laughter while the man looks around helplessly. She looks at him through the window, gives a tiny satisfied smile, then casually turns away. Ultra-realistic Korean subway lighting, gritty DV-inspired look, natural motion blur, realistic physics, handheld imperfections, authentic reactions and ambient subway audio; no music. No gore, blood, serious injuries, identity drift, outfit changes, duplicated people, teleportation, regenerating glass, distorted hands, CGI look, subtitles, captions, text, logos or watermark.

video prompts · #video-prompts
prompt
sample output for "photorealistic portrait prompt"

photorealistic portrait prompt

Use the attached photo as the identity reference. Preserve the person's facial features, face shape, eye shape, nose, lips, skin texture, body proportions, hairstyle (unless otherwise specified), age, ethnicity, and overall identity exactly. Do not alter facial geometry or make the person look like someone else. Create a photorealistic editorial portrait of the person standing in an English cottage garden with foxgloves in bloom, captured during golden hour with warm rim lighting, shallow depth of field, and a natural background bokeh. The person has a soft, natural smile and a relaxed, confident pose while looking toward the camera. The image should appear as if photographed on a Hasselblad with an 80mm lens, featuring realistic skin tones, accurate lighting, fine detail, subtle 35mm film grain, premium color grading, and a luxury fashion/editorial aesthetic. Maintain the person's exact identity while enhancing only the photography quality, lighting, and overall cinematic presentation. No facial reshaping, beautification that changes identity, or artificial-looking skin.

image generation · #image-generation
mindwiki · a shelf, not a feedsubmit a find →