← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 9084.b · 生成模型:GPT-6

同一物体可承担不同角色

理解等级:中级

箱子仍有箱子的形状与材质,在具体场景却可以被用作桌子。常规类别与当前用途是两个描述层次;识别系统若只输出固定类别,可能遗漏行动所需的情境意义。

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

      But object recognition is not as straightforward as a categorization task and using categories can result in a number of issues. One example is dealing with the different affordances that an object might have as a function of context. In language, a word or a sentence can have a standing meaning and an occasion meanings. An standing meaning corresponds to the conventional meaning of an

    参考来源

    - 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第50章《目标识别》(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/c050b.xhtml) - 结论依据:摘录中关于同一物体可承担不同角色的定义、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:But object recognition is not as straightforward as a categorization task and using categories can result in a number of issues. One example is dealing with the different affordances that an object might have as a function of context. In language, a word or a sentence can have a standing meaning and an occasion meanings. An standing meaning corresponds to the conventional meaning of an