GPT Image 2.5 第一人称包内视角人像提示词
上传照片做面部参考,生成从毛绒包内仰视的第一人称视角人像,并叠加 iOS 来电界面,附完整提示词。

Gpt image 2.5 on chatgpt Prompt ⤵️ Use my uploaded photo as the face reference. 3:4 vertical. Point-of-view shot from inside a grey fleece-lined bag looking up. The soft fuzzy lining frames every edge of the image. the person in reference, keeping her exact face, hairstyle and identity, reaches down into the bag toward the camera, her fingers foreshortened in the foreground and stacked with chunky silver rings. She wears a black ribbed beanie, black oval sunglasses, a navy blazer and a blue-and-white striped shirt: a white London townhouse with a dark grey-green front door numbered 27, black iron railings and a small olive tree. Overlaid iOS incoming call interface in white system font: a "Message" button on the left and a "Voicemail" button on the right as frosted round icons, and at the bottom a frosted glass pill slider reading "slide to answer" with a white circle holding a green phone icon. Natural daylight, candid street-style fashion photography, realistic skin texture. Make it realistic Improve its colour grading
这条是引用着另一条写的:「GPT Image 2.5 上下分割编辑海报提示词」