← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 9014.b · 生成模型:GPT-6

平移可以统一成矩阵乘法

理解等级:中级

列向量约定下,二维平移矩阵 T=[[1,0,t_x],[0,1,t_y],[0,0,1]],作用于 [x,y,1]^T 得到 [x+t_x,y+t_y,1]^T。平移在原坐标需加法,在齐次坐标与旋转等操作统一为乘法;全局非零缩放 T 不改变其射影映射。

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

      What is interesting is that we have transformed an addition into a product by a matrix, T, and a vector, p, in homogeneous coordinates. Rewriting the equation above as: (38.10) with the homogeneous matrix, T, defined as (38.11) For calculations in homogeneous coordinates, both matrices and vectors are only defined up to an arbitrary (non-zero) multiplicative scaling factor. To cascade two translation operations t and

    参考来源

    - 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第38章《图像与几何表示》(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/c038b.xhtml) - 结论依据:摘录中关于平移可以统一成矩阵乘法的定义、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:What is interesting is that we have transformed an addition into a product by a matrix, T, and a vector, p, in homogeneous coordinates. Rewriting the equation above as: (38.10) with the homogeneous matrix, T, defined as (38.11) For calculations in homogeneous coordinates, both matrices and vectors are only defined up to an arbitrary (non-zero) multiplicative scaling factor. To cascade two translation operations t and