Smalltown Boy AI ダンス：二人の表情と役割交代

Edit the complete @Video1. @Image1 replaces ONLY the person who first performs the exaggerated close-up at the lens. @Image2 replaces ONLY the person who initially dances behind them and later takes the close-up. These are two independent identity references, never start/end frames. Preserve each photo subject’s face, hair, body appearance, clothes and accessories in every shot. Track identities through the foreground/background role exchange; never swap identities when positions change. Preserve the outdoor setting, wide-angle close-ups, camera movement, action timing, role exchange, side-by-side dance and complete final approach from @Video1. Do not shorten the sequence or replace it with a generic dance. Keep coherent hands, feet and separate bodies. Hold the final pose only for the sub-second rounding remainder. Silent output, with no existing recording or named voice.

送信時に追加指示と音声設定を反映します。参考による生成はフレーム単位の一致を保証しません。