How the reference pictures align with the target video… | Deepfake AI видео

Промпт (текст в видео), по которому сгенерирован этот ролик

How the reference pictures align with the target video - Picture 1 is the exact opening frame at 0.00 seconds. Picture 2 is the one-way lowered-hand ending frame at 14.00 seconds. Use no additional references. integrated_multimodal_description: [Shot 1] Static locked-off vertical 9:16 anime-illustration shot. Keep the opening image's exact girl, upward left-facing profile, face, hair, pale-blue blouse, red-and-black bow, dark skirt, color, linework, sky, sea, petals, background planes, and foreground blue-plane blur. Keep the camera completely still. This is a fast one-way wind tunnel, not a loop and not a floating tableau. All background planes are already in a fast constant lateral stream at frame 0. Every visible plane translates screen-right to screen-left in a straight, committed path, covering a large visible distance; planes enter from the right edge, cross behind and around the girl, and exit the left edge. New planes continually replace the exiting planes from the right edge. Their speed is brisk and uniform, with only slight forward flutter of paper edges; no local bobbing, no pendulum swings, no pause, no oscillation, no reversal, no orbit, and no hovering. Pink petals are carried in the same fast right-to-left current. Hair and blouse stream steadily left with the wind. 0.00-1.20 seconds: the raised hand makes two tiny aiming motions while holding the blue hero plane; the whole background current keeps moving fast. 1.20-1.80 seconds: she performs a single sharp forward wrist flick and releases it, blinking once precisely on release. 1.80-2.60 seconds: the hero plane accelerates immediately nose-first along its existing point direction into the open upper-left sky, traveling at high speed on a straight forward trajectory and exiting the frame through the left boundary. It must never circle, curve around the girl, slow near her, reverse, or move toward her face, wrist, shoulder, or body. 2.60-4.20 seconds: the hero plane remains off-screen; the background planes and petals maintain fast right-to-left transit. The empty hand smoothly tucks one hair strand near her temple; head, torso, gaze, and expression do not move. 4.20-6.00 seconds: hold the settled hair-tuck pose while the fast plane and petal stream continues its one-way right-to-left transit at unchanged speed. 6.00-8.20 seconds: the same hand travels down once in a clear smooth natural path along the viewer-left side of her body and comes to rest relaxed by her lower torso, matching Picture 2; this is the only later body action. 8.20-14.00 seconds: hold the lowered-hand pose with the exact same calm upward gaze and expression while the fast plane and petal stream continues its one-way right-to-left transit, ending exactly on Picture 2. No camera movement, new character angle, extra limbs, second blink, text, realism, or style drift. overall_soundscape: Strong clean wind rush, rapid light paper flutter, distant open-air ambience. non_diegetic_music: N/A