구글 AI 아바타, 대본만 주면 알아서 연기해요
2026.06.19 · 785 readers


챗지피티 이미지 2로 다음 프롬프트로 내 사진을 변환해줘.
프롬프트:
Keep the face and identity of the person in the reference image — preserve their facial features, eyes, nose, lips, and skin tone naturally. Natural body proportions, realistic head-to-body ratio. Do not enlarge the face.
photograph. Early 2000s compact CCD digital camera snapshot. Cool-toned low saturation, muted blue-green shadows, soft digital noise, slightly blown highlights in sunlit areas, flat compressed dynamic range. Candid hand-held framing, slightly unsteady, point-and-shoot 35mm equivalent lens, auto mode, no compression artifacts.
Bright summer afternoon, warm natural sunlight filtering through dense overhead campsis clusters with dappled light on the subject, occasionally a little overexposed in the brightest patches. Relaxed candid moment with a soft, gentle smile, eye-level angle. Vary the framing and pose naturally each time within a candid range — sometimes the subject placed on the left or right third, sometimes centered half-body, sometimes a looser full-body, with loose unposed body language; do not lock to one composition.
Minimal, barely-there makeup: natural dewy skin, soft tinted lip, natural brows, no heavy coverage.
Outfit: a light, airy summer garment in a dry matte fabric, in a bright cool-leaning summery palette — vary it freely each time and do not fix one specific item or color (e.g. linen sundress, spaghetti-strap slip dress, sleeveless linen shirt, light cotton blouse, breezy camisole, in soft white / sky-blue / cream / pastel tones), relaxed flowy fit, sometimes slightly oversized.
Background: an old stone brick wall completely covered in climbing campsis radicans / campsis grandiflora (nungsohwa) vines — dense clusters of vivid orange-red trumpet-shaped flowers and lush dark green leaves, a few fallen petals scattered on the ground, some petals softly out of focus in the foreground. Bright summer daytime outdoor scene.
No text, no watermark, no 3D render, no enlarged head, no disproportionate face size. No studio lighting, no heavy makeup, no modern sharp digital sensor.
Was this helpful? Share with your friends
Share
Get daily AI trend updates