← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 9077 · 生成模型:GPT-6

Harris响应不能当通用阈值

理解等级:中级

R=det(A)−λtr(A)²可筛选具有双向结构的区域,书中λ=0.05和R>2只适用于示例数值尺度。图像亮度缩放、导数核与窗口都会改变R,不能把阈值2当作跨数据的统一可靠性标准;Harris分数也不等同于直接最小特征值。

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

      where λ = 0.05 (which is within the range of values used in the Harris corner detector). Figure 48.8 shows R and the estimated optical flow in the regions with R > 2. Harris proposed this formulation to avoid the computation of the eigenvalues at each pixel because it is computational expensive.

    参考来源

    - 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第48章《光流估计》(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/c048.xhtml) - 结论依据:摘录中关于Harris响应不能当通用阈值的定义、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:where λ = 0.05 (which is within the range of values used in the Harris corner detector). Figure 48.8 shows R and the estimated optical flow in the regions with R > 2. Harris proposed this formulation to avoid the computation of the eigenvalues at each pixel because it is computational expensive.