Cinematic AR Visualization in Indian Street Market
grey t-shirt, green cargo pants, white sneakers. standing, holding phone. use uploaded face 100% accuracy. A hyper-realistic, cinematic 8k image of a man standing in a bustling Indian street market at night.
Full prompt
{
"A hyper-realistic, cinematic 8k image of a man standing in a bustling Indian street market at night. He is wearing a plain grey t-shirt, green cargo pants, and white sneakers, holding a smartphone. The scene is overlaid with a detailed Augmented Reality (AR) visualization: numerous floating, translucent frosted-glass music player interface cards (resembling Spotify/Apple Music UI) orbiting the subject in 3D space. The cards feature colorful album artwork, progress bars, and playback controls with a soft, glowing border luminescence (bloom effect). The composition uses depth of field, with some cards blurred in the foreground and others sharp near the subject. The background features authentic Indian street elements: green and yellow auto-rickshaws, shop signs with Hindi text, and warm ambient street lighting contrasting with the cool blue-white glow of the AR interface.",
"negative_prompt": "coat, blazer, suit, formal wear, cartoon, illustration, low quality, blurry, distorted, bad anatomy, extra limbs, text watermark",
"subject_attributes": {
"clothing": "grey t-shirt, green cargo pants, white sneakers",
"pose": "standing, holding phone",
"facial_reference": "use uploaded face 100% accuracy"
},
"environment": {
"location": "Indian street market",
"time": "night",
"details": "auto-rickshaws, crowds, shop lights"
},
"style": "Augmented Reality, Cyberpunk, Cinematic, Photorealistic"
}
“Don’t hesitate, move forward with mayo.”
A close-up, 1:1 aspect-ratio image of freshly cooked, piping hot {argument name="dish_name_en" default="okonomiyaki"} on a griddle.
Make this look like a poorly taken snapshot shot at a place with latitude and longitude {argument name="latitude and longitude" default="35.31721720077321"}.
Using the attached image as a motif, we generated differential images of the face of a woman slightly intoxicated in an izakaya, consisting of four scenes divided into a 2x2 grid.