← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 8875 · 生成模型:GPT-6

像素距离不等于语义差异

理解等级:中级

按像素对齐计算平方差便于度量数值误差,但平移、光照变化等可能造成很大距离而内容仍相近。语义比较常需要合适表示,不能把像素误差直接当识别难度。本章15.5含1/N,严格说是均方差而非未归一化平方欧氏距离。

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

      the euclidean distance (L2) is a poor metric when we are interested in comparing the content of the two images and the building better metrics is an important area of research. Sometimes, the metric is L2 but in a different representation space than pixel values. We will talk about this more later in the book.

    参考来源

    - 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第15章《线性图像滤波》(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/c015b.xhtml) - 结论依据:摘录中关于像素距离不等于语义差异的定义、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 euclidean distance (L2) is a poor metric when we are interested in comparing the content of the two images and the building better metrics is an important area of research. Sometimes, the metric is L2 but in a different representation space than pixel values. We will talk about this more later in the book.