Photogpt
INPUT A = Style Reference (press-surrounded editorial scene) INPUT B = Identity Reference (person to preserve) STYLE TRANSFER RULE - Replicate camera density, microphone arrangement, framing pressure, and composition from INPUT A - Do not copy the face from INPUT A - Apply style only, preserve identity from INPUT B IDENTITY LOCK (FROM INPUT B) - Preserve exact facial structure - Preserve proportions and symmetry - Preserve eye shape and spacing - Preserve nose and lip shape - Preserve skin tone - Preserve hairstyle - Preserve upper body outfit - Maintain natural neutral expression - Do not modify facial geometry FRAMING - Chest-up portrait - Subject centered - Eye-level perspective - 50mm lens compression - Vertical composition - Tight crop with equipment filling frame edges PRESS ENVIRONMENT (FROM INPUT A STYLE) - Surround subject with dense modern broadcast cameras - Professional video cameras aimed directly at subject - Large shotgun microphones entering from all directions - Foam microphone heads overlapping heavily in foreground - Equipment occupying majority of outer frame - Some cameras partially obscure frame edges - Face remains fully visible and unobstructed CAMERA SCREENS - Multiple visible LCD camera screens in foreground - Screens display live preview of INPUT B’s face - Accurate screen perspective and brightness - Realistic color temperature shift on screens LIGHTING - Balanced editorial lighting similar to INPUT A - Even facial exposure - Soft shadow under chin - Controlled highlights on camera bodies - Realistic reflections on lenses and glass - No paparazzi flash bursts BACKGROUND - Dark neutral press background - No logos - No unrelated objects - Clean equipment-only environment MATERIAL REALISM - High-detail microphone foam texture - Realistic plastic and metal camera surfaces - Accurate lens reflections - Natural skin texture - No stylization - No cartoon rendering CONSTRAINTS - No distortion of INPUT B’s face - No exaggerated emotion - No vintage cameras - No motion blur - No cinematic color grading - No AI-art texture artifacts

PhotoGPTのAI画像

Gemini Nano Banana Pro
プロンプト

INPUT A = Style Reference (press-surrounded editorial scene) INPUT B = Identity Reference (person to preserve) STYLE TRANSFER RULE - Replicate camera density, microphone arrangement, framing pressure, and composition from INPUT A - Do not copy the face from INPUT A - Apply style only, preserve identity from INPUT B IDENTITY LOCK (FROM INPUT B) - Preserve exact facial structure - Preserve proportions and symmetry -…

サイズ928px x 1152px
参照画像
元画像 1
元画像 2

創作を始める

数千人のアクティブユーザーが集まるDiscordコミュニティに参加して、最新のプロンプトアイデアやコミュニティ作品をチェックしましょう

画像を生成Discordに参加
Google Veo 3.1Seedance 2

関連作品

コミュニティの似たAI生成作品を探索できます

このスタイルを試す