在网上学到一个用 ChatGPT 最新的图像制作功能来实现连续像素图动画(分割图)的方法。
提示词是:
Create a detailed pixel art frame animation for a game, where the final image is divided into multiple sub-images, each serving as a continuous animation keyframe. Design the sequence to depict [(这里填入你想要生成的像素图的内容)] Ensure the keyframes transition smoothly and continuously, and include as many frames as possible to achieve a high level of fluidity and detail in the animation.
我实际尝试填入的是:
Create a detailed pixel art frame animation for a game, where the final image is divided into multiple sub-images, each serving as a continuous animation keyframe. Design the sequence to depict [ChatGPT LOGO的机器人从手机屏幕里钻出来] Ensure the keyframes transition smoothly and continuously, and include as many frames as possible to achieve a high level of fluidity and detail in the animation.
于是就得到了:
这个方法是从 X 上的一位 ID 名为 Majid Manzarpour
的用户那里看来的,但我也不清楚他是否就是最初的原创者。不过我也看到评论说,虽说肉眼看起来已经算是不错的连续像素图了,但实际上用起来可能还有许多需要修改、编辑的地方,才能真正用于游戏制作。
此外,我也尝试了直接用中文指示生成单张(非动画)的点阵像素图。比如这样:
这张图的提示词是:创建一张 8bit 颜色的明显具有像素图风格的图,严格遵守 8bit颜色。颜色要醒目、跳跃、诡异和瑰丽、对比度强烈。创造一张高科技上海外滩风景的像素图。背景夜空上要出现巨大的“魔都”两个字。看起来很雄伟和诡异。横向图。