← 返回起始笔记位置

知识卡片

普通读书笔记卡 · 8899.b · 生成模型:GPT-6

匀速平移模型的假设

理解等级:中级

全局平移序列写成 ℓ(x,y,t)=ℓ₀(x−vₓt,y−vᵧt)。它假设速度恒定且随物体移动的亮度不变,连续坐标允许亚像素速度。相机曝光改变、光照变化、形变和遮挡都可能破坏该模型;它是理解局部视频的基线,并非一般动态场景的精确表达。

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

      One special sequence is when the image has a global motion with constant velocity (vx, vy). In such a case we can write: (19.1) where ℓ0(x, y) = ℓ(x, y, 0) is the image being translated, and vx and vy are constants. At time t the image ℓ0(x, y) is translated by the vector (vxt, vyt) as described by equation (19.1). The pixel value at

    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:摘录中关于匀速平移模型的假设的定义、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:One special sequence is when the image has a global motion with constant velocity (vx, vy). In such a case we can write: (19.1) where ℓ0(x, y) = ℓ(x, y, 0) is the image being translated, and vx and vy are constants. At time t the image ℓ0(x, y) is translated by the vector (vxt, vyt) as described by equation (19.1). The pixel value at