← 返回起始笔记位置

知识卡片

普通读书笔记卡 · 9072.e · 生成模型:GPT-6

前进与变焦的差异在深度

理解等级:中级

前进的径向扩张率为V_Z/Z,变焦的扩张率为ḟ/f。前者随深度变化,后者对全图相同;但正对同深度平面时二者可能难以区分。观察多个深度层次与已知相机参数,可减少仅凭扩张外观的误判。

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

      Before moving into motion estimation, let's consider one final scenario: a static camera observing a static scene, but the focal length changes over time. What will the motion field be? In this setting, despite that there is not motion in the scene, the focal length of the camera changes over time, producing motion in the image plane. If the focal length increases, it will seem

    参考来源

    - 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第47章《三维运动及其二维投影》(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/c047.xhtml) - 结论依据:摘录中关于前进与变焦的差异在深度的定义、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:Before moving into motion estimation, let's consider one final scenario: a static camera observing a static scene, but the focal length changes over time. What will the motion field be? In this setting, despite that there is not motion in the scene, the focal length of the camera changes over time, producing motion in the image plane. If the focal length increases, it will seem