← 返回起始笔记位置

知识卡片

普通读书笔记卡 · 8901.d · 生成模型:GPT-6

运动匹配模糊沿轨迹平均

理解等级:中级

把时空高斯改为 gᵥ(x,y,t)=g(x−vₓt,y−vᵧt,t),相当于沿目标速度的轨迹聚合信息。与该速度匹配的物体保留较多清晰结构,其他速度与静态背景则更容易模糊;仍存在原核的空间平滑,不能称为完全无模糊。该变换是剪切而非简单旋转。

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

      How could we create a filter that keeps sharp objects that move at some velocity (vx, vy) while blurring the rest? Figure 19.4(c) shows the desired output of such a filter. The bottom image shows one frame for a sequence filtered with a kernel that keeps sharp objects moving left at 1 pixel/frame while blurring the rest. This filter can be obtained by skewing the

    参考来源

    - 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第19章《时间滤波器》(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/c019.xhtml) - 结论依据:摘录中关于运动匹配模糊沿轨迹平均的定义、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:How could we create a filter that keeps sharp objects that move at some velocity (vx, vy) while blurring the rest? Figure 19.4(c) shows the desired output of such a filter. The bottom image shows one frame for a sequence filtered with a kernel that keeps sharp objects moving left at 1 pixel/frame while blurring the rest. This filter can be obtained by skewing the