🎬 Loading...
    Prompt

    I2VA: For the target video, at 0.00 seconds into the target video, <Picture 1> is fully referenced as the exact starting frame and character identity. SUBJECT: Exact same man from Picture 1 — short brown hair, beard, dark jacket, jeans, sitting at the café table in the train station. Maintain perfect facial and body likeness. ACTION / TIMING (9–10 seconds total): [0.0–2.5s] He sits normally at the table looking down at his sandwich and coffee in regular gravity. [2.5–10.0s] Zero gravity suddenly activates. His hair and jacket fabric begin to float upward. The sandwich, coffee cup, juice, newspaper, backpack, and chair start rising and drifting. He looks around in panic, eyes wide, and says: “What the hell?!” while flailing slightly as everything floats around him. End on him panicking amid the floating objects. ENVIRONMENT / LOOK: Exact busy train station from Picture 1 — departure boards, other travelers, café tables — now in zero gravity with objects floating. CAMERA: Starts matching the medium reference framing. Holds steady so the sudden loss of gravity, the floating objects, and his panicked reaction are clearly visible. AUDIO (native stereo): - Soft station ambience at the start - Quiet whoosh as zero gravity begins - Soft floating / drifting sounds of objects and fabric - Clear male voice, panicked: “What the hell?!” - Quiet station tone - No music PRESERVE: Exact face, clothing, and likeness from Picture 1. Keep only the objects already in the image floating. Show clear panic on his face and in his movement. AVOID: Inventing new objects or making him stay calm.

    Generation Settings

    Parameters used to generate this content

    Created with
    ComfyUI
    Sampler
    res_multistep
    Seed65111541922843
    Steps8
    ComfyUI Workflow

    The node graph used to generate this content — pan and zoom to explore, or download it to run in ComfyUI.

    Info
    Video
    Likes
    4
    Created
    9/22/2026
    Creator
    crafted101
    Source
    CivitAI
    Models Used

    AI models used to generate this content

    Actions