知识卡片

普通读书笔记卡 · 9014.a · 生成模型:GPT-6

齐次点相加依赖代表选择

理解等级:中级

p 与 λp 是同一个齐次点,但直接计算 p+q 与 λp+q 通常得出不同的点,因此任意齐次代表向量相加不是良定义的点运算。若需要有限点的中点,应先归一化到 w=1 再做仿射组合,不能把向量加法直接当作射影点运算。

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

      It is important to mention that while you can add two points in heterogeneous coordinates, you can not do the same in homogeneous coordinates! Using homogeneous coordinates, we can place a point at infinity by setting w = 0. This is called the ideal point. 38.3 2D Image Transformations One important operation in computer vision (and in computer graphics) is geometric transformations of shapes. Some

    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:摘录中关于齐次点相加依赖代表选择的定义、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:It is important to mention that while you can add two points in heterogeneous coordinates, you can not do the same in homogeneous coordinates! Using homogeneous coordinates, we can place a point at infinity by setting w = 0. This is called the ideal point. 38.3 2D Image Transformations One important operation in computer vision (and in computer graphics) is geometric transformations of shapes. Some