
원본 문서
GPT Image 2 · en
{
"type": "Cinematic 3-Frame Sequence Stills",
"style": "Hyperrealistic, emotional, natural light, shallow depth of field, low-saturation earth tones, soft overcast atmosphere, delicate film grain.",
"subject": {
"count": 1,
"description": "A young female traveler, with dark brown hair tied in a messy bun, wearing a dark oversized sweater, brown trousers, and hiking boots.",
"face_visibility": "All three frames show either a profile or a back view."
},
"layout": {
"orientation": "A vertical triptych, consisting of three horizontally stacked panels separated by a thin border.",
"sections": [
{
"title": "Top frame",
"position": "top",
"count": 1,
"labels": [
"The woman leaned against the wooden fence with her arms crossed, looking to the right."
]
},
{
"title": "medium format",
"position": "center",
"count": 1,
"labels": [
"A woman sat alone in the middle of a deserted, winding road, holding a folded paper map, looking to her left."
]
},
{
"title": "Bottom frame",
"position": "bottom",
"count": 1,
"labels": [
"A woman's back view, carrying a rugged hiking backpack, standing and gazing at the lake and valley."
]
}
]
},
"environment": {
"location": "A remote highland landscape, with rolling hills, sparse dead trees, heather and shrubs, distant mountains and lakes.",
"weather": "Cloudy, cool, with light fog"
},
"camera": {
"framing": "The film's narrative sequence is as follows: the top frame shows a medium shot of a person, the middle frame shows a full-body seated pose, and the bottom frame shows a medium shot of a person's back.",
"lens": "shallow focus documentary photography style",
"aspect": "High vertical composition"
},
"details": {
"props_count": 3,
"props": [
"The wooden fence has a small circular trail marker with a white house icon.",
"Folded paper map",
"Brown canvas hiking backpack"
]
}
}사용 방법
- 매개변수를 포함한 전체 원본 프롬프트를 복사하세요.
- 주제, 문구, 구도를 자신의 요구에 맞게 바꾸세요.
- 호환되는 외부 도구에서 사용하고 이용 조건과 결과를 확인하세요.
시작 전 확인
- 참조 이미지가 필요한지 확인하세요. 호환 표시는 해당 모델로 이 예시를 생성했다는 뜻이 아닙니다.
- 이름, 수치, 이미지 속 문구를 확인하세요. 모델 버전에 따라 결과가 다를 수 있습니다.





