知识卡片

普通读书笔记卡 · 9075.d · 生成模型:GPT-6

局部恒速把多点方程合并

理解等级:中级

Lucas–Kanade在邻域内假设共享一个(u,v),用高斯权重累加各点梯度残差平方。纹理在多个方向变化时可补足单点孔径歧义;窗口跨越不同运动时共享速度假设失效,即使线性系统可解也可能得到错误混合。

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 equation constrains the motion (u, v) in location (x, y) to be along a line perpendicular to the image gradient ∇ℓ = ℓx, ℓy at that location. This is the same relationship that we discussed before when describing the aperture problem. This is not enough to estimate motion and we will need to add additional constraints. The second assumption that we will add is

    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:摘录中关于局部恒速把多点方程合并的定义、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 equation constrains the motion (u, v) in location (x, y) to be along a line perpendicular to the image gradient ∇ℓ = ℓx, ℓy at that location. This is the same relationship that we discussed before when describing the aperture problem. This is not enough to estimate motion and we will need to add additional constraints. The second assumption that we will add is