Vertical Candid Portrait of Woman
GPT Image 2.5 image prompt by 王大锤 · original post
A prompt for generating a vertical candid photograph of a young East Asian woman with specific features.
the prompt
Vertical 9:16 candid photograph of an adult East Asian woman in her mid twenties with very fair natural skin, a slim tall figure, long slightly messy black hair, soft oval face, clear dark eyes and glossy natural pink lips, kneeling in a shallow riverside stream while slowly leaning forward and letting both hands sink into the water, her gaze drifting just past the camera with a quiet slightly dazed expression. She occupies about 52% of the frame height and is cropped around mid thigh, with her head positioned about 24% below the top edge; the camera is only about 32cm above the water surface, lower than her waist, tilted upward by about 8 degrees, photographed with a 50mm lens. A wet tree branch about 18cm from the lens enters from the upper left and covers about 22% of the frame, overlapping the upper edge of her hair and left shoulder, its shape still recognizable but strongly defocused; a second patch of blurred leaves covers about 15% of the lower left corner and lightly cuts across her left forearm. The main light comes from the 4 o’clock direction at roughly 22 degrees low backlight, hard late afternoon sunlight broken by leaves, catching only the outer edge of her hair, cheekbone, shoulder and wet forearms while the front of her face stays softly filled by reflected water light. Deep green #31483a occupies about 42% of the image, pale river silver #b9c6c7 about 25%, warm ivory about 18%, natural skin tone about 12%, with the remaining area formed by small white water highlights. She wears a pale ivory lightweight summer dress with thin straps and soft gathered fabric, the lower fabric partially soaked and clinging naturally around the thighs with irregular wet creases, plus a slim black metal choker; skin keeps visible pores, tiny water droplets and slight unevenness. Focus falls precisely on the eyes and face, the foreground branch melts into a soft dark green shape, the river behind becomes medium-sized sparkling bokeh, and only the moving tips of her hair and fingertips show slight motion blur while her face remains sharp. Three fingerprint details must remain visible: one narrow wet strand of hair sticks diagonally across her right collarbone; a single pale leaf is caught against the submerged hem near her left thigh; one small patch of moss on the dark rock behind her forms an irregular crescent shape. Avoid symmetrical centered framing, fully unobstructed subject, plastic airbrushed skin, both eyes staring directly into the lens, studio softbox lighting, uniform global blur, duplicated jewelry, warped hands.
- model
- GPT Image 2.5
- type
- image
- shape
- 2:3
- prompt
- 425 words
Made this? Fix the credit or remove it.










