GPT Image 2AI-промпт
Tekken Style Fight Storyboard Grid — Sora 2 Prompt
Generate a cinematic 3x3 fighting game storyboard with consistent characters, full HUD, and AAA Tekken-inspired visuals using two reference images.
ПРОМПТ · EN
{ "instruction": "使用提供的两张角色图像作为精确的视觉参考。完美保留身份、面部、服装、比例和细节。", "concept": "argument name=\"game concept\" default=\"以 Tekken 风格游戏网格故事板呈现的 AAA 格斗游戏场景\"", "layout": { "type": "3x3 网格", "description": "九个分镜展示连续的战斗过程，每帧均可见完整的游戏 UI" }, "panels": [ "分镜 1：戏剧性的 VS 画面 — 角色 A 对战角色 B，两人面对面，发光的背景，中心显示 versus 文字", "分镜 2：回合开始 — 两位格斗家处于待机姿态，屏幕显示 READY 文字", "分镜 3：FIGHT 时刻 — 战斗开始时电影般的火花爆炸", "分镜 4：首次攻击 — 角色 A 出拳，带有动态模糊", "分镜 5：反击 — 角色 B 出脚，带有冲击火花", "分镜 6：连招序列 — 多重打击火花、运动轨迹、动态镜头", "分镜 7：低血量时刻 — 血条闪烁红色，紧张时刻", "分镜 8：终结技 — 慢动作重击，电影级视角", "分镜 9：KO 画面 — 对手向后倒下，巨大的 KO 文字，粒子飞溅" ], "style": "argument name=\"art style\" default=\"超写实 AAA Tekken 风格格斗游戏，电影级游戏渲染，高细节角色，写实光影\"", "camera": "游戏内格斗视角，略带角度的侧视图，冲击时动态缩放", "environment": "高端格斗竞技场，带有霓虹点缀的深色背景，微妙的景深，不喧宾夺主", "ui_elements": { "health_bars": "左上和右上，细节丰富，发光，随分镜逐渐减少", "timer": "顶部中心", "player_names": "左上和右上", "round_indicator": "可见", "combo_counter": "在打击时出现", "hit_effects": "火花、闪光、光束", "KO_text": "大号粗体，最终分镜屏幕中心", "READY_FIGHT_text": "街机风格粗体排版" }, "effects": [ "攻击时的动态模糊", "冲击火花", "能量轨迹", "镜头抖动效果", "终结打击时的慢动作感", "打击时的屏幕光晕" ], "color_grading": "argument name=\"color grading\" default=\"高对比度，街机电影感，鲜艳的高光，深邃的阴影\"", "output": "整洁的 3x3 网格，完美对齐的分镜，完整的 UI 可见，无 Logo，无水印"
}
"reference_style": "Tekken 8 游戏 UI，街机格斗游戏界面，现代 AAA 格斗 HUD"О промпте
Generate a cinematic 3x3 fighting game storyboard with consistent characters, full HUD, and AAA Tekken-inspired visuals using two reference images. Use it as a Концепт-арт starting point for GPT Image 2: keep the visual structure and style constraints intact, then swap in your own subject, brand, or scene.
Start with small changes to subject, scene, camera, and color, then generate a few variations to compare composition and detail without turning the page copy into keyword stuffing.
Как использовать этот промпт
- Copy the entire prompt JSON block into your AI video generator.
- Choose a model that supports image references and multi-frame outputs (e.g. Sora 2 or compatible).
- Replace the two character reference images with your own portraits to maintain identity across all nine panels.
- Adjust the
{argument}variables (game concept, art style, color grading) to match your creative vision. - Generate and review the 3x3 grid, ensuring all UI elements and fight progression are visible.
