OpenZhenZhen AI プロンプトギャラリー · gpt-image-2.5 · image

朝のラッシュ時に地下鉄で通勤するアジア人女性のスマホスナップ

Candid Phone Shot of an Asian Woman Commuting on the Subway at Morning Rush Hour

「朝のラッシュ時に地下鉄で通勤するアジア人女性のスマホスナップ」は gpt-image-2.5 を使用した作品です。OpenZhenZhen で元のプロンプトと生成例を確認し、創作の参考にできます。

朝のラッシュ時に地下鉄で通勤するアジア人女性のスマホスナップ — AI image creation
朝のラッシュ時に地下鉄で通勤するアジア人女性のスマホスナップ

プロンプト

中国語プロンプト

3:4竖幅单张真实手机后置摄像头通勤抓拍;仅参考上传参考图中明确成年亚洲女性的五官、脸型、黑色长直发与自然肤质,保持外貌一致,不进行换脸或五官漂移,完全重建场景。清晨早高峰,她站在拥挤的城市地铁车厢内,穿浅灰无袖针织上衣、深灰宽松长裙、平底鞋,肩背浅色帆布包;一只手抓住顶部吊环,另一只手低头握手机,身体随列车晃动微微侧倾,认真看手机,神情专注并带一点刚睡醒的困倦,嘴唇自然闭合,不看镜头、不刻意微笑。长发自然垂在肩胸前,少量碎发散开,衣物和帆布包有真实褶皱与受力。周围有许多正常通勤乘客,有人抓扶手、看手机或背双肩包,顺着车厢空间自然站立,不围住主体、不全部看镜头;前景一两位乘客的肩膀或背包可轻微遮挡边缘,但不遮脸和关键动作。背景包含车门、吊环、扶手、路线图、冷白顶灯和模糊广告;不要有正对镜头的粗大竖直铁柱,也不要遮挡人物。手机屏幕模糊,不出现清晰文字、聊天记录或品牌。普通手机约26mm广角,从斜前方一两米处拍摄,中近景,画面略倾斜,有轻微车厢晃动与自然运动模糊,但面部清晰;保留真实肤质、轻微眼下疲惫、碎发、衣物纹理和早高峰杂乱感。负向提示:空车厢、乘客太少、独自站立、模特摆拍、商业摄影、直视镜头微笑、所有乘客看镜头、乘客围住主体、身体穿模或挤压变形、重叠手臂、多余肢体或手指、错误抓环姿势、悬空吊环、手机变形、清晰手机文字、铁柱挡脸或居中、方向混乱、过度磨皮、强美颜、背景严重虚化、换脸、五官漂移、文字、水印。

英語プロンプト

3:4 vertical single-frame candid commute snapshot from a normal rear phone camera. Use the uploaded reference only to keep the clearly adult Asian woman's facial features, face shape, long straight black hair, and natural skin texture consistent; do not face-swap or drift her features, and fully reconstruct the scene. During early-morning rush hour she stands inside a crowded urban subway car, wearing a light-gray sleeveless knit top, loose dark-gray long skirt, flat shoes, and a light canvas shoulder bag. One hand naturally grips an overhead ring while the other holds a phone down by her side; her body tilts slightly with the swaying train as she seriously reads the phone, focused with a hint of just-awake fatigue, lips naturally closed, not looking at the camera and not deliberately smiling. Her long hair falls naturally over her shoulders and chest with a few stray strands; clothing and bag show believable wrinkles and tension. Many ordinary commuters surround the space naturally: some hold rails, look at phones, or wear backpacks; they do not encircle the subject and are not all looking at the camera. One or two foreground shoulders or backpacks may slightly occlude the frame edges to convey crowding, but never cover her face or key actions. Show a regular subway carriage with doors, overhead rings, handrails, route maps, cool-white ceiling lights, and blurred advertisements. No thick vertical iron pillar directly facing or blocking her. Keep the phone screen blurry with no readable text, chat records, or brand. Shoot diagonally from one or two meters away with an approximately 26mm wide angle and medium-close framing, as if a friend casually photographed her on the same train; slightly tilted frame, subtle carriage sway and natural motion blur, while her face remains clear. Retain real skin texture, slight under-eye fatigue, flyaway hair, clothing wrinkles, and the cluttered rush-hour atmosphere. Negative prompts: empty carriage, too few passengers, standing alone, model posing, commercial photography, direct smiling eye contact, everyone looking at camera, commuters encircling the subject, clipping or crushing bodies, overlapping arms, extra limbs or fingers, wrong ring-gripping posture, floating rings, deformed phone, readable phone text, iron pillar blocking or centered on the face, chaotic directions, excessive smoothing, strong beauty filter, heavily blurred background, face swap, facial-feature drift, text, watermark.

API モデルを見る

画像・動画プロンプトをもっと見る · 画像を生成 · 動画を生成 · AI キャンバスで作成