← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 9032.e · 生成模型:GPT-6

平面视差在校正条件下仿射

理解等级:中级

在校正针孔双目、固定基线和焦距等条件下,三维平面使逆深度对像素坐标为仿射,因而视差的二阶空间导数为零。可据此偏好分段平面,而非直接偏好深度二阶为零;边界、曲面和一般未校正视差不满足同一公式。

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

      Another interesting property of disparity is that for flat surfaces, the second-order derivative along any spatial direction, x, is zero: (40.22) This equality does not hold for depth. This property allows introducing a simple regularization term that favors planarity. To translate the returned disparity for a given pixel d [n, m] into depth values z [n, m], we need to invert the disparity, that is

    参考来源

    - 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第40章《立体视觉》(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/c040.xhtml) - 结论依据:摘录中关于平面视差在校正条件下仿射的定义、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:Another interesting property of disparity is that for flat surfaces, the second-order derivative along any spatial direction, x, is zero: (40.22) This equality does not hold for depth. This property allows introducing a simple regularization term that favors planarity. To translate the returned disparity for a given pixel d [n, m] into depth values z [n, m], we need to invert the disparity, that is