知识卡片

普通读书笔记卡 · 9071.b · 生成模型:GPT-6

瞬时刚体流可分两部分

理解等级:高级

静态刚性场景的相机运动流可写为深度相关的平移项和不依赖深度的旋转项,两者由同一像点坐标和标定控制。书中称只适用小位移,是针对有限帧近似;瞬时微分表达本身不要求角速度数值很小,只需导数存在。

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

      This expression can be rewritten as, using x = fX/Z and y = fY/Z: (47.23) This is the expression we were looking for. It relates the 2D motion field with the camera velocity and rotation. The matrices are only a function of the intrinsic camera parameters (focal length, f) and the camera coordinates. Note that this expression is only valid for small displacements. In the

    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:摘录中关于瞬时刚体流可分两部分的定义、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:This expression can be rewritten as, using x = fX/Z and y = fY/Z: (47.23) This is the expression we were looking for. It relates the 2D motion field with the camera velocity and rotation. The matrices are only a function of the intrinsic camera parameters (focal length, f) and the camera coordinates. Note that this expression is only valid for small displacements. In the