GPT Image Prompt
Using the active LoRA identity reference image(s) as the absolute identity authority, create a realistic vertical iPhone-style smartphone photo of the exact same real woman, preserving her full face and body identity without beautification or reshaping. Keep her recognizable facial geometry, natural eye spacing, nose-lip-jaw relationship, warm skin tone, short dark bob haircut, same shoulder width, same torso length, same torso-to-leg balance, same waist thickness, same hip-to-thigh transition, same leg length ratio, and the same casual physical presence. Do not slim, lengthen, narrow, idealize, or replace her body with a more model-like version. She must still read unmistakably as the exact same woman from the active LoRA identity reference, with both face and body continuity preserved.
She is standing on an old cobblestone village street, slightly leaning forward under the believable weight of a large dark hiking backpack. One hand grips the backpack strap near her chest with realistic hand tension and natural finger structure, while the other hand hangs lower holding a small plastic water bottle with believable grip alignment. Keep the white fitted sleeveless tank top, olive green cargo shorts with worn pocket details, loose red neck scarf, small hoop earrings, layered necklace, visible backpack straps, and the small flag patch on the backpack. Her short dark hair keeps its natural volume and shape, with a few loose strands falling across her face from outdoor air.
Her expression must stay subtle, alive, and human: a brief camera-aware glance, relaxed eyelids, lips slightly parted naturally, calm travel-worn softness in her face, not a frozen stare, not a posed smile, not mannequin-like, and not influencer-pretty. Preserve the source-prompt human moment while keeping the LoRA facial structure intact. The feeling should be that she has been walking and briefly looks toward the phone, slightly tired but present, grounded, and real.
The environment is a physically believable old village alley with weathered stone buildings, shutters, patches of greenery, textured cobblestones, and ordinary lived-in depth. Keep the scale measurable and real, with realistic human-to-environment proportions, believable perspective, authentic street depth, and natural spatial imperfection. The woman must feel physically grounded on the cobblestones with realistic weight placement, contact shadows, and no floating or composited appearance.
All expression, gesture, outfit, pose, backpack weight behavior, environment, camera framing, lighting, and realism treatment remain subordinate to the active LoRA identity reference image(s). The final image must still be unmistakably the exact same LoRA subject, not a similar-looking substitute, not a beautified replacement, and not a generic attractive travel woman. Face identity and body geometry from the LoRA reference remain the authority at all times.
Capture it as a casual travel photo taken by someone nearby on an iPhone, not a professional portrait. Use vertical smartphone framing with a close crop, slightly uneven composition, mild handheld imbalance, loose natural framing, and believable smartphone wide-angle behavior without flattering distortion that changes body proportions. Keep it photographed, not staged. Use natural daylight with realistic smartphone exposure behavior, subtle highlight clipping if needed, realistic skin texture, localized softness, mild compression, slight reconstruction inconsistencies, small-sensor depth response, and ordinary background depth without cinematic isolation.
Favor believable smartphone realism over visual perfection: slight asymmetry, imperfect crop decisions, mild environmental clutter, realistic texture inconsistency, subtle processing artifacts, and everyday social-photo energy. Hands must have correct anatomy and believable finger spacing. Clothing must respond naturally to gravity and body posture. Backpack weight should affect posture and strap tension realistically. Avoid cinematic lighting, fashion-editorial posing, fake HDR polish, exaggerated blur, oversharpening, hyper-clean rendering, luxury travel aesthetics, influencer-style beautification, over-symmetry, dead-eyed blankness, copied reference stiffness, or any perspective that slims the waist, elongates the legs, narrows the hips, or changes the reference body identity. Realism always wins over beauty.
How to use: Copy and paste this prompt into ChatGPT, Gemini, Nano Banana, or your favorite AI image tool. If it mentions LoRA, reference image, uploaded photo, or identity reference, upload your own photo first. No reference photo? Replace the character with a clear age, gender, and nationality, such as “25-year-old Japanese woman” or “30-year-old American man”.