一位年輕日本女性的身份參考，保留她的面部結構、膚色、髮型特徵、表情神態以及具辨識度的外觀。創作一張寫實的室內直式智慧型手機自拍照，從極低的地面角度向上拍攝。這名日本女孩站在畫面下方中央，穿著休閒的寬鬆 T 恤，低頭看著鏡頭，帶著調皮的嘟嘴和放鬆的社群媒體自拍表情。在畫面正上方，一隻倒立的蓬鬆貓咪向鏡頭傾斜，直接向下看，以可愛、搞笑、迷人的表情主導上方構圖，與女孩溫柔友善的氛圍相得益彰。明亮現代的公寓室內，大窗戶，柔和的自然日光，乾淨極簡的背景，寫實的皮膚與毛髮質感，自然的智慧型手機照片光線，下方為女孩、上方為倒立貓咪的平衡構圖，休閒隨性的氛圍，可愛、獨特、適合社群媒體、照片般寫實、高細節。反向提示詞：平視相機、平淡自拍角度、側面位置的動物、隨機動物、卡通、動漫、CGI、扭曲的解剖結構、多餘的肢體、雜亂的房間、可怕的表情、具攻擊性的動物行為、模糊、低解析度、浮水印、標誌、文字。
@saniaspeaks_
View original source與俏皮貓咪和日本女性的低角度自拍
一張寫實的低角度智慧型手機自拍照，捕捉了一位年輕的日本女性，以及一隻倒立探入鏡頭畫面中的可愛蓬鬆貓咪。
125
Likes
3.8K
Views
10
Reposts
71
Comments
40
Saves
How to use this prompt
- 1Copy the prompt with the copy button on the prompt panel.
- 2Upload your reference image — this prompt transforms an input image.
- 3Click “Generate in Sunra”, or paste the prompt into GPT Image 2.
- 4Tweak the subject, style, or details and regenerate until it’s right.
More GPT Image 2 prompts
About GPT Image 2
GPT Image 2 is OpenAI’s image model, strong at following detailed, multi-part instructions and rendering legible in-image text, with reliable prompt adherence for complex scenes.
Explore GPT Image 2 promptsFrequently asked questions
- Which model is this prompt for?
- This prompt is written for GPT Image 2. Run it in Sunra, or paste it into GPT Image 2 directly.
- Do I need a reference image?
- Yes — this prompt edits or transforms an uploaded image, so add a reference image when you generate.
- How do I get the best results?
- Keep the prompt’s structure, then adjust the subject, style, and details to fit your idea. Regenerate a few times and pick the best output.
- Is the prompt free to use?
- Yes — copy the prompt for free. Hit “Generate in Sunra” to create images with it instantly.
Make it your own
Run this GPT Image 2 prompt in Sunra and generate your own version in seconds.
Generate in Sunra