AI Elf Dance：用自己的照片制作圣诞精灵舞蹈贺卡

@Image2 is an original three-panel visual guide, read LEFT to RIGHT: greeting, grounded side step, small quarter turn. Use its Christmas room, costume and full-body framing only. Ignore the illustrated adult's face, hair and body identity; replace that person with @Image1 throughout.
These panels are states of ONE subject over time, not three subjects. Synthesize continuous motion; never show the guide image, panel borders, stills or a slideshow. The sample video displayed on the page is NOT a model input.
Create a new 8-second photorealistic vertical 9:16 Christmas greeting from @Image1, the ONLY adult identity source. Preserve the adult's face, skin tone, hair, age appearance and natural adult body proportions. Replace everyday clothes with ONE consistent festive outfit: green velvet elf tunic, red collar, black belt, opaque red trousers, green curled-toe shoes and a green-and-red elf cap. Do not shrink the body or enlarge the head.
A warm living room with one lit Christmas tree, wrapped gifts and a fireplace. Single static frontal FULL-BODY shot from start to finish. Keep the entire hat and both shoes visible, with floor space under the feet and space above the hat. Empty hands; no handheld objects or additional people.
0–1.5s: face camera, smile and give one small open-hand greeting at chest level, feet planted.
1.5–4s: two gentle side steps, one to screen-left and one back to center, with small shoulder bounces and relaxed bent arms. Transfer weight naturally; the supporting foot stays in contact with the floor. No high jumps.
4–6.5s: one slow quarter turn to screen-right, then return to face camera using two grounded steps. This is a simple adapted greeting dance, not a request for frame-exact choreography.
6.5–8s: settle both feet, raise open hands to shoulder height and hold a friendly final greeting for half a second.
Keep the same face, outfit, shoes, hat and room through the turn. Real body and foot movement, never a static portrait zoom or slideshow. No sliding feet, extra limbs, facial morphing, camera cuts, zoom, text, logos or watermarks. Include the original audio directed below.
Generate an ORIGINAL instrumental festive soundtrack at about 110 BPM: light sleigh-bell shakers, warm plucked strings, rounded bass and soft handclaps. Cheerful and cozy, no famous Christmas melody, no recognizable song quotation, no lyrics or vocals. A short upbeat entrance accompanies the wave; keep one coherent groove through the side steps and quarter turn, with quiet foot taps when feet contact the floor. Resolve with a gentle bell accent at the final greeting. Keep the music comfortable and the room ambience subtle. No speech or narration.

把照片中的熟悉人物放进精灵服装，制作短小的圣诞舞蹈或问候视频。本页以一位成年人和全身客厅场景开始。

可以尝试原创单人精灵问候。本工具独立运行，不调用 ElfYourself、不提供它的固定舞步，也不把多张脸映射到现成群舞视频。

本配方只提供一张成年人照片。双人身份和宠物舞步尚未验证，因此本页不把它们作为可用模式。

不需要，照片提供面孔和身份，提示词请求精灵服。衣服会有意替换，但面部细节与肢体仍可能变化。

编辑提示词免费；生成按页面显示的积分计费。包含原创生成节日配乐，不提供指定歌曲。已完成的视频仍计费；确认失败的任务按现有流程退回积分。

上方视频是使用虚构主体制作的原创 AI 生成示例。分镜图是独立的可选模型输入。一个示例不能保证你的照片具有相同的动作、节奏或相似度。