← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 9020.b · 生成模型:GPT-6

正交投影忽略深度缩放

理解等级:中级

正交投影可用 [[1,0,0,0],[0,1,0,0],[0,0,0,1]] 把 [X,Y,Z,1]^T 映为 [X,Y,1]^T。它不按 Z 缩小远处物体,与透视模型有不同假设。将平行投影结果与透视结果比较前,先确认所用相机模型和坐标尺度。

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 can use homogeneous coordinates to describe a wide variety of camera models. For instance, the projection matrix for an orthographic projection (equation [5.10]), K, is (39.4) as can be verified by multiplying by a 3D point in homogeneous coordinates. 39.3 Camera-Intrinsic Parameters Now that we have seen how 3D to 2D projection can be effectively modeled using homogeneous coordinates and simple vector computations, let's

    参考来源

    - 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第39章《相机建模与标定》(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/c039.xhtml) - 结论依据:摘录中关于正交投影忽略深度缩放的定义、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 can use homogeneous coordinates to describe a wide variety of camera models. For instance, the projection matrix for an orthographic projection (equation [5.10]), K, is (39.4) as can be verified by multiplying by a 3D point in homogeneous coordinates. 39.3 Camera-Intrinsic Parameters Now that we have seen how 3D to 2D projection can be effectively modeled using homogeneous coordinates and simple vector computations, let's