← 返回起始笔记位置

知识卡片

普通读书笔记卡 · 8978.e · 生成模型:GPT-6

分层视差近似空间深度

理解等级:中级

多平面动画把场景绘制在前后不同层上,使前景与背景按不同速度移动,产生视差深度感。视觉分层表示也可用有序平面近似部分三维结构,减少完整三维建模复杂度。但平面近似不能表达所有曲面、视角变化和层内遮挡细节。

这张卡片看不懂?

已有反馈 · 0 条

    还没有人提交反馈。

    在整书笔记中的位置

    原书出处 · 1 处

    • · 查看原书依据

      Animators for these cartoons used a device called a multiplane camera, which worked as follows. A camera looked at a stack of glass planes. On each plane, the animators would draw a different layer of the scene. For example, on the back plane would be the sky, then there would be mountains on the next plane, then trees, and then the foreground characters. During filming

    参考来源

    - 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第31章《知觉分组》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)/_epub-src/OPS/xhtml/c031.xhtml) - 结论依据:摘录中关于分层视差近似空间深度的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。 - 原始内容:Animators for these cartoons used a device called a multiplane camera, which worked as follows. A camera looked at a stack of glass planes. On each plane, the animators would draw a different layer of the scene. For example, on the back plane would be the sky, then there would be mountains on the next plane, then trees, and then the foreground characters. During filming