← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 8955.a · 生成模型:GPT-6

马尔可夫性是条件独立

理解等级:中级

图像MRF的直觉是:给定某像素的足够邻域后,该像素与其余位置条件独立。这不意味着邻域之外原本无关,也不意味着邻域一给定就能唯一确定中心值。边缘分布、条件分布和条件独立分别描述不同性质。

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

      A powerful class of image models is known as Markov random field (MRF) models. They have a precise mathematical form, which we will describe later, in chapter 29. For now, we present the intuition for these models: if the values of enough pixels surrounding a given pixel are specified, then the probability distribution of the given pixel is independent of the other pixels in the

    参考来源

    - 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第27章《统计图像模型》(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/c027b.xhtml) - 结论依据:摘录中关于马尔可夫性是条件独立的定义、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:A powerful class of image models is known as Markov random field (MRF) models. They have a precise mathematical form, which we will describe later, in chapter 29. For now, we present the intuition for these models: if the values of enough pixels surrounding a given pixel are specified, then the probability distribution of the given pixel is independent of the other pixels in the