电影感女性海盗船长
GPT Image 2 提示词GPT Image2提示词GPT Image 2 提示词案例AI提示词案例AI图片生成提示词案例AI绘画提示词案例OpenAI图像生成提示词ChatGPT画图提示词
Full GPT Image 2 Prompt
Create a cinematic photorealistic full-body portrait of {argument name="character name" default="a female pirate captain"} standing confidently on the wet wooden deck of a 17th-century sailing ship at sea, centered in frame and facing the viewer, with her face deliberately obscured by a soft rectangular blur for anonymity. She has {argument name="hair color" default="long dark purple-black hair"} tied in a high messy topknot with loose strands, and wears an ornate layered pirate outfit: a black embroidered long coat with gold trim, fitted dark corset, cream ruffled blouse cuffs, asymmetrical tattered lace skirt, burgundy fabric panels, multiple belts, pouches, buckles, hanging charms, a small pocket watch, thigh-high slit exposing one leg, and tall brown leather gladiator-style boots. She holds one long flintlock pistol angled downward in her right hand. The setting is a dramatic pirate ship deck under bright daylight: blue ocean horizon, white clouds, rigging ropes, sails, mast hardware, wooden railings, coils of rope, barrels, crates, and polished rain-slick planks reflecting sunlight. Include exactly 4 background crew members: 1 crouched or leaning sailor firing a smoking cannon on the left, 1 distant standing sailor near the rigging behind the captain, 1 sailor working ropes beside barrels on the right, and 1 sleeveless crew member partially visible at the far right with a blurred face. Include exactly 1 visible cannon on the left emitting thick white smoke, exactly 2 prominent rope coils on the right, and exactly 2 wooden barrels near the right-middle background. Use {argument name="visual style" default="ultra-detailed cinematic photorealism"}, natural sunlight from the upper left, high contrast, shallow depth of field, realistic fabric texture, weathered wood, brass and leather details, sea mist, and an adventurous swashbuckling mood. Square 1:1 composition, full-body subject, low eye-level perspective, no text, no watermark.