← 返回起始笔记位置

知识卡片

普通读书笔记卡 · 8976.a · 生成模型:GPT-6

可命名性不是分组前提

理解等级:中级

语言中的物体和关系帮助说明结构化表示的价值,但视觉上连贯的形状不必已有名称。两个陌生物体仍可能被分开感知,因此知觉组不能只定义为词典中的对象类别,也不能把图像转文本看作保留所有视觉结构的唯一方式。

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

      Although the analogy to text makes clear why perceptual organization is important, perceptual groups can also go beyond the limits of language. Some structures we see are hard to name but still clearly represent a coherent group in our heads. Consider, for example, the image in figure 31.2 below.

    参考来源

    - 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第31章《知觉分组》(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/c031.xhtml) - 结论依据:摘录中关于可命名性不是分组前提的定义、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:Although the analogy to text makes clear why perceptual organization is important, perceptual groups can also go beyond the limits of language. Some structures we see are hard to name but still clearly represent a coherent group in our heads. Consider, for example, the image in figure 31.2 below.