知识卡片
可转向金字塔分解尺度方向
steerable金字塔在每层用方向带通滤波器产生子带,并让低通支路继续下采样递归。每个子带可描述特定尺度的某类方向结构。它比各向同性细节残差更细分,也带来更多系数;子带响应是表示而非对输入图像平移完全不变的常数特征。
这张卡片看不懂?
这张卡片已通过人工审查,不再接受失败反馈。
已有反馈 · 0 条
还没有人提交反馈。
在整书笔记中的位置
- · 图像金字塔
原书出处 · 1 处
- · 查看原书依据
The Laplacian pyramid provides a richer representation than the Gaussian pyramid. But we would like to have an even more expressive image representation. The steerable pyramid [441] adds information about image orientation. Therefore, the steerable representation is a multiscale oriented representation that is translation-invariant. It is non-aliased and self-invertible. Ideally, we'd like to have an image transformation that was shiftable, that is, where we could
参考来源
- 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第23章《图像金字塔》(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/c023.xhtml)
- 结论依据:摘录中关于可转向金字塔分解尺度方向的定义、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:The Laplacian pyramid provides a richer representation than the Gaussian pyramid. But we would like to have an even more expressive image representation. The steerable pyramid [441] adds information about image orientation. Therefore, the steerable representation is a multiscale oriented representation that is translation-invariant. It is non-aliased and self-invertible. Ideally, we'd like to have an image transformation that was shiftable, that is, where we could