ทำซีรีส์แนวตั้งด้วย AI
แบบล็อกหน้าตัวละครทั้งเรื่อง

KAEW: a young rural Siamese woman aged 19 with a youthful smooth face, slim and strong, honey-tan skin,
oval face with clear cheekbones, narrow dark eyes, dark eyebrows, one small mole just below the outer
corner of her LEFT eye, short black hair cropped at ear length and swept back; she wears a dull indigo
cotton breast cloth wrapped crosswise over the chest and tied behind the neck, a faded mangosteen-red
wrap trouser, barefoot, no jewelry; she carries a small old brass water bowl with one dent on its rim.

Imagine a vertical 9:16 photorealistic cinematic film still.
[โทนเรื่องและยุคสมัย หนึ่งย่อหน้า ใช้ซ้ำทุกช็อต]
[บรรยายช็อตนี้: ขนาดภาพ มุมกล้อง ใครอยู่ตรงไหน ทำอะไรอยู่]
Setting: [สถานที่]. [แสงของช็อตนี้]
[วางสเปกตัวละครจากขั้น 1 ของทุกตัวที่อยู่ในช็อต]
[NAME] matches the attached character reference exactly: same face, age, skin tone, hair, body and
clothes. Never redesign, never add jewelry, shoes or modern items.
Only [NAME] is in the foreground. No other people are in the frame.
Anatomically correct, physically plausible, realistic skin, no text, no watermark.
Avoid: [สิ่งที่ไม่ต้องการ เช่น modern clothes, text, cartoon, extra fingers, different character faces]

Create a VIDEO (a 10-second moving clip with motion and natural sound, NOT a still image). Vertical 9:16.
This video is a direct visual continuation of the FIRST attached image: treat it as frame zero of the shot.
Preserve exact character identity, faces, clothing, hair, body positions, prop placement, camera position,
lighting and location exactly as in that image, and continue the motion naturally from that moment.
Other attached images are identity references only.
ACTION: [อะไรขยับ ใครทำอะไร เรียงตามลำดับ]
CAMERA LOCK: unless the ACTION explicitly describes a camera move, the camera is locked off on a tripod:
no zoom, no reframing, no cut to another angle; it is ONE continuous shot with no edit.
PEOPLE LOCK: only the people visible in the first frame exist; never add, duplicate, replace or remove
a main character.
No on-screen text, no subtitles, no music, no narration.
Sound: [เสียงในฉาก]
END FRAME TARGET: [เฟรมสุดท้ายต้องเห็นอะไร]. Do not complete any extra action after that end frame.