知识卡片

普通读书笔记卡 · 8913.b · 生成模型:GPT-6

插零扩展是抽取的转置

理解等级:中级

整数扩展把原样本放到 k 的倍数位置,其余位置置零,矩阵为相同尺寸约定下抽取矩阵的转置。它保留输入而未插补未知值,仍需后续滤波。转置是代数关系,不能误称抽取的逆:插零不能找回此前被丢弃的像素。

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

      Expansion by a factor k consists in increasing the number of samples in the image by inserting k – 1 zeros between each sample on the horizontal dimension, and then on the vertical dimension. This will give us the image of size Nk × Mk. We will use the notation:

    参考来源

    - 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第21章《图像下采样与上采样》(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/c021.xhtml) - 结论依据:摘录中关于插零扩展是抽取的转置的定义、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:Expansion by a factor k consists in increasing the number of samples in the image by inserting k – 1 zeros between each sample on the horizontal dimension, and then on the vertical dimension. This will give us the image of size Nk × Mk. We will use the notation: