
A completely natural, photorealistic smartphone selfie of a young East Asian woman casually walking along a quiet tree-lined street, captured as if she quickly took the photo herself with her phone while walking. She has long, slightly messy dark black hair falling naturally over her shoulders, with a few loose strands blown across her face by a light breeze. She has minimal makeup, realistic skin texture, subtle under-eye detail, natural lips, and a relaxed slightly open-mouth expression while looking directly into the phone camera. She is wearing a soft lavender-purple knitted sweater with a slightly loose fit, a brown shoulder bag strap crossing naturally over her shoulder, and a dark floral skirt barely visible at the bottom of the frame. Her arm is extended toward the camera in an obvious handheld selfie position, creating natural perspective and slight wide-angle distortion. The background shows an ordinary residential road surrounded by mature green trees, parked cars, pavement, and patches of sunlight filtering through the leaves. Bright midday natural light creates uneven highlights and shadows across her face and sweater. Make it look like a real unedited phone photo rather than a professional portrait: imperfect framing, slightly tilted composition, mild motion softness, realistic smartphone exposure, subtle lens flare, natural dynamic range, tiny hair flyaways, authentic skin pores, ordinary background details, no cinematic color grading, no studio lighting, no beauty filter, no artificial sharpness, no DSLR look. Vertical smartphone photo, casual everyday snapshot, extremely realistic and believable.
This image was generated with GPT Image 2 by OpenAI. The prompt on this page is the exact text used to create it.
The example was generated with GPT Image 2 using: 3:4 · Character · Street Photography · Realistic. Reusing these settings gives the closest match.
Copy the prompt as-is first, then replace only the subject while keeping the style, lighting, and composition keywords. Generate a few variations — outputs differ between runs.
Yes. Prompts work best as starting points — adjust descriptors, aspect ratio, or style tags to make the output your own.