Cyber City Girl Group — An MV-Style Entrance

Model N/AVerticalSci-FiFashionNeon NightCamera Movement15s
A Michael Bay-style entrance mashed up with a K-pop MV: a rain-soaked night street in a futuristic cyber-industrial city, an arcing camera introducing each member in turn, variety-show text cards popping up behind them. A complete second-by-second shot list.

PROMPT · Prompt

Create with AI
Michael Bay-style entrance + 'Love's Theme Song' Season 2~ This time we went with a Korean-style futuristic cyber city vibe~ Prompt: The nighttime setting is locked to a futuristic cyber industrial city at night, an empty street after rain, no people, no other vehicles. The environment is locked to fully match the reference image (if any), live-action realism, cinematic-grade CG effects, trendy Korean girl group MV style, shallow depth of field, cool-warm color grading. Text on title cards keeps the poster's style (if any), appearing randomly on screen behind the girl group members and between the city background, like a variety show effect. Natural ambient light source. 0-2 seconds: medium shot, slowly moving in a leftward arc. Girl group member 1 stands under a streetlamp, looking up at the camera. Focus is locked on girl group member 1. 2-5 seconds: the camera continues moving in a leftward arc, focus shifting forward. In the foreground, the blurred silhouette of girl group member 2 gradually comes into focus, becoming a low-angle close-up, turning toward the camera. Blurs into bokeh. 5-9 seconds: the camera pulls back, panning left. Retreats to a medium shot. Girl group member 3 walks in from the left, occupying the foreground. Focus shifts from member 2 to member 3. 9-12 seconds: the camera moves in an orbiting motion. Pulls back. Girl group member 4 walks in from off-screen right, becoming a foreground close-up. 12-15 seconds: wide-angle lens, the camera pulls back and rises, then stabilizes. The four girl group members strike a powerful stance — walking to the left of the frame and stopping to look at the camera, positioned on both sides, close together. The street and thick fog create depth of field, the wet ground reflects light; from the thick fog at the center background, a vehicle slowly drives out toward the camera. Strictly avoid clipping/mesh intersection throughout, no music in the entire film, no subtitles, no dialogue, the 4 girl group members must have no similar appearance, no identical hairstyles, no identical demeanor, no identical outfits. 8:44am · August 22, 2026 · 3,315 views 3 3 43 22 Related

✍️ Editor’s Notes

This one runs on a shot-by-shot timeline plus explicit camera-path tracking — a new shot every 2 to 4 seconds shifts focus from member 1 through member 4, and the camera-movement description gets down to vector-level specificity like “arc left” and “orbit around,” far more precise than just saying the camerawork should look cool. The closing line — “the four members must have no similar faces, no matching hairstyles” — is a preemptive anti-face-cloning prompt for multi-person generation. The biggest trap in a multi-character entrance scene is the model getting lazy and stamping every supporting face out of the same mold, and this prompt inoculates against that ahead of time. No model is specified, presumably left for a stronger model capable of handling multi-subject shot composition to take on.

Create with AI

Related Works

׋›
↑