Prompt

Create an ultra-realistic cinematic aerial shot that seamlessly transitions from the provided start frame to the end frame. Set high above New York City on the antenna of the Empire State Building, matching the composition, lighting, scale, and perspective of both reference frames precisely. A black flag flutters continuously in strong wind with realistic, heavy cloth physics, occasionally stretching fully before folding back. The two people remain calm with only subtle posture shifts and natural clothing movement. The camera behaves like a stabilized cinema drone, performing one smooth clockwise orbit while slowly pulling back and ascending—no FPV motion, shake, or whip pans. Gradually reveal more of the Empire State Building and Manhattan, with realistic atmospheric haze, subtle river movement, distant traffic, and natural lighting. End by perfectly matching the final reference frame and hold for one second as the flag continues to flutter. The result should be indistinguishable from

Result

Gemini Omni Image to Video

Settings

Reference images