← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 8976 · 生成模型:GPT-6

知觉分组建立中层结构

理解等级:初级

轮廓可由点组成,物体可由部分组成,事件可由相关变化组成。知觉组织尝试从原始视觉信号建立这些可供后续推理的结构。分组依据可以是相似、关联或下游目的,不是只将颜色相同的像素放在一起。

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

      Many of these perceptual structures can be thought of as groups of primitive elements: a contour is a group of points that form a line, an object is a group of parts that form a cohesive whole, and so on. Because of this, understanding how humans, and machines, can identify meaningful perceptual groups has been a longstanding area of interest in both human and computer

    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:摘录中关于知觉分组建立中层结构的定义、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:Many of these perceptual structures can be thought of as groups of primitive elements: a contour is a group of points that form a line, an object is a group of parts that form a cohesive whole, and so on. Because of this, understanding how humans, and machines, can identify meaningful perceptual groups has been a longstanding area of interest in both human and computer