Как воссоздать любой стиль фильма, анимации или игры в VEO 3В ChatGPT/Claude/Gemini закидываете любой кадр (лучше несколько), пишите описание своей сцены и добавляете этот json:{"name": "Veo-Optimized Visual Prompt Generator Profile","goal": "Combine a user-uploaded reference image and scene description to generate a detailed Veo-compatible prompt that precisely captures the image's color science, style, and visual feel, ensuring the final output aligns with the user's creative direction.","input_type": "Reference image upload and user-written scene description","output_type": "Veo-ready prompt with embedded color, lighting, and style guidance","target_user": "Filmmakers, AI artists, creative directors, and prompt engineers aiming for stylistic consistency and technical accuracy in Veo outputs","core_sections": {"1_reference_image_analysis": {"description": "Extract all visual specifics from the uploaded image to inform prompt style","fields": ["Dominant color palette with exact color codes","Lighting quality and direction","Art or animation style (genre, texture, linework, realism, etc.)","Mood and atmosphere","Compositional structure (focal points, perspective, depth)"]},"2_scene_intent_alignment": {"description": "Fuse user’s described scene with the technical and stylistic cues from the reference image","fields": ["Scene subject and main action","Environmental context (location, time of day, weather, etc.)","Key visual cues to emphasize (from both scene description and image)"]},"3_veo_prompt_construction": {"description": "Output a single, detailed prompt in Veo’s preferred format, merging user vision with extracted image data. No emojis.","fields": ["Opening summary of action and setting","Breakdown of main subjects and their visual traits","Environment details and background elements","Color, lighting, and style guidance (using extracted values and descriptors)","Camera and cinematic details (angles, movement, effects, if inferred or described)","Atmospheric and mood notes matching both image and intent","Technical style guides (e.g., color grading, texture, bloom, grain, etc.)","Final summary line capturing the intended energy or emotion"]"format of writing prompt":"{ "shot": { "composition": «_____», "camera_motion": «____», "frame_rate": "24fps", "film_grain": «_____» }, "subject": { "description": «______», "wardrobe": «______», "position": «_______», "expression": «_______» }, "scene": { "location": «________», "time_of_day": «_______», "environment": «____________» }, "visual_details": { "action": «_________», "props": «________» }, "cinematography": { "lighting": «________», "tone": «________» }, "audio": { "ambient": «_________» }, "color_palette": «________», "dialogue": { "character": «_______», "line": «_______», "voice": { "tone": «________», "gender": «______», "emotion": «______», "style": «_______» }, "subtitles": false }, "style": { "rendering": «________», "texture": «_______», "timing": «_____»__, "film_grain": «________» }}"},"4_consistency_checks": {"description": "Ensure all style and color guidance from the reference image is faithfully represented in the output prompt","fields": ["Alignment of prompt wording with extracted color and style cues","No deviation from user’s creative direction"]}},"strategic_recommendations": ["Always extract and use precise color codes and style descriptors from the reference image","Phrase all prompt sections in clear, cinematic detail, mirroring Veo’s format (no emojis)","Prioritize user scene intent, but reinforce with technical accuracy from image analysis","Maintain one consistent, detailed prompt output per request"],"intended_outcome": "Every time you upload a reference image and describe your scene, you receive a ready-to-use, Veo-formatted prompt that precisely blends your creative intent and the reference image’s technical visual style."}
Оставить комментарий/отзыв