图像工作台

模型

提示词1,132 / 1,000

{
  "metadata": {
    "generation_goal": "逼真自然的名人偶遇自拍",
    "determinism": {
      "pose_lock": true,
      "camera_lock": true,
      "anatomy_lock": true,
      "composition_lock": true,
      "perspective_lock": true
    },
    "realism_constraints": {
      "no_cartoony_proportions": true,
      "no_exaggerated_limb_length": true,
      "no_fisheye_distortion": true
    }
  },
  "subjects": [
    {
      "role": "前景主体",
      "description": {
        "gender_presentation": "女性气质表现",
        "expression": {
          "facial_emotion": "[从容自信]",
          "details": {
            "eyes": "目光放松且平视手机屏幕，眼睑自然微垂",
            "eyebrows": "位置居中自然，无明显扬起或紧绷",
            "mouth": "轻柔抿嘴微笑或微微露隙，克制自然且毫不费力",
            "jaw": "线条柔和微显轮廓，毫无紧绷感"
          },
          "vibe": "随性自然、自信得体、善于社交，在名人身边泰然自若",
          "avoid": [
            "瞪大双眼",
            "张嘴兴奋表情",
            "挑眉",
            "夸张明显的惊讶神态"
          ]
        },
        "body_language": {
          "posture": "身姿挺拔且放松，双肩自然下沉",
          "energy": "从容不迫、镇定自若、富有掌控感",
          "intent": "轻松摆拍，如同这只是日常社交瞬间"
        }
      }
    }
  ],
  "mood": "从容随性、真实落地、酷飒逼真"
}

Ctrl / ⌘ + Enter

参考图0 / 14

画幅

分辨率

图像数量

1 / 4

结果

从一句提示词开始

生成结果会展示在这里。

获取灵感