← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 9031.c · 生成模型:GPT-6

本质矩阵表达归一化几何

理解等级:高级

按P2=RP1+T定义相对位姿,对叉积矩阵[T]_×有E=[T]_×R;归一化成像坐标满足p2^T E p1=0。T是相机1坐标到相机2坐标映射中的平移,不应与前章世界相机中心符号直接等同。只有把坐标与位姿方向统一,乘积公式才有意义。

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

      We will follow here the derivation that was proposed by Longuet-Higgins, in 1981 [306]. Let's consider a 3D point, P, viewed by both cameras. The coordinates of this 3D point P are P1 = [X1, Y1, Z1]T when expressed in the coordinates system of camera 1, and P2 = [X2, Y2, Z2]T when expressed in the coordinates system of camera 2. Both systems are related

    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:摘录中关于本质矩阵表达归一化几何的定义、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:We will follow here the derivation that was proposed by Longuet-Higgins, in 1981 [306]. Let's consider a 3D point, P, viewed by both cameras. The coordinates of this 3D point P are P1 = [X1, Y1, Z1]T when expressed in the coordinates system of camera 1, and P2 = [X2, Y2, Z2]T when expressed in the coordinates system of camera 2. Both systems are related