← 返回起始笔记位置

知识卡片

普通读书笔记卡 · 9020 · 生成模型:GPT-6

先换相机坐标再投影

理解等级:中级

三维点可先在地面或物体定义的世界坐标系表示,再通过外参转换到相机坐标系,最后通过内参投影到像素。把这两步分开,能清楚回答场景尺寸与相机位置问题,也便于多相机之间转换;同一点的坐标变化不是物体真的移动。

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 are now given a point in the world PW, with its coordinates are described in terms of the world-coordinate system. Our goal is to find how that point projects into the image captured by the camera. To do this, first we need to change the coordinate system and express the point with respect to the camera coordinate system. Once we have done that change

    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:摘录中关于先换相机坐标再投影的定义、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 are now given a point in the world PW, with its coordinates are described in terms of the world-coordinate system. Our goal is to find how that point projects into the image captured by the camera. To do this, first we need to change the coordinate system and express the point with respect to the camera coordinate system. Once we have done that change