OpenZhenZhen AI 프롬프트 갤러리 · gpt-image-2.5 · image

주황빛 황무지의 오리지널 생존자 영화 스틸

Original Survivor Movie Still in the Orange Wasteland

‘주황빛 황무지의 오리지널 생존자 영화 스틸’은(는) gpt-image-2.5로 만든 작품입니다. OpenZhenZhen에서 원본 프롬프트와 생성 예시를 살펴보고 새로운 창작에 활용해 보세요.

주황빛 황무지의 오리지널 생존자 영화 스틸 — AI image creation
주황빛 황무지의 오리지널 생존자 영화 스틸

프롬프트

중국어 프롬프트

创作一幅原创实拍感末世电影静帧,画面只有一位面部身份与 ref-10a-v1 一致的成年荒原幸存者。使用 ref-10a-v1 作为唯一主体的身份参考,保留其成年身份与自然面部结构,不做美化。采用四分之三身像,躯干侧转、头部转向镜头。幸存者双手在腰部安全地向下持握一把明显为非功能性道具的自制霰弹枪;不瞄准、不射击,画面不出现攻击、伤口或血迹。主体穿磨损皮革外套、围巾和沾有沙尘的实用装备。场景设在开阔的橙色沙地荒原,头顶为饱和青蓝色天空,远处有漂浮沙尘和轻微虚化、无品牌标识的追击车辆。只在远处沙尘中加入克制的热浪与轻微动态柔化,主体保持清晰对焦。采用强烈的沙漠顶光和边缘坚硬的深阴影,表现可触的沙粒、磨损金属、织物纹理及自然的汗尘细节。使用电影感青橙色调,同时保持自然肤色与克制的模拟胶片质感;不承诺具体胶片型号或相机规格。整体原始、烈日炙烤、紧张且原创,不出现可识别的系列服装、角色、标志、文字、水印或其他人物。

영어 프롬프트

Create an original live-action post-apocalyptic cinematic still featuring one adult wasteland survivor whose facial identity matches ref-10a-v1. Use ref-10a-v1 as the identity reference for the sole subject, preserving the adult identity and natural facial structure without beautification. Show a three-quarter-body portrait with the torso angled sideways and the head turned toward camera. The survivor holds a clearly nonfunctional improvised shotgun prop safely downward at waist level with both hands; it is not aimed or fired, and the scene shows no attack, injury, or blood. Dress the subject in distressed leather outerwear, a neck scarf, and dusty practical gear. Set the still in an open orange-sand desert under a saturated turquoise sky, with drifting dust and a softly blurred, unbranded pursuit vehicle far behind. Add subtle heat haze and restrained motion softness in the distant dust only, while keeping the subject sharply focused. Use harsh overhead desert sunlight, deep hard-edged shadows, tactile sand, worn metal, fabric weave, and natural sweat-and-dust detail. Apply a cinematic teal-and-orange palette with realistic skin tones and a restrained analog-film character; do not claim a specific film stock or camera specification. Keep the image raw, sun-blasted, tense, and original, with no recognizable franchise costumes, characters, logos, text, watermark, or additional people.

API 모델 보기

더 많은 이미지·동영상 프롬프트 보기 · 이미지 생성 · 동영상 생성 · AI 캔버스에서 만들기