← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 9195.c · 生成模型:GPT-6

过滤条件可能删除真正目标

理解等级:中级

按簇大小或几何筛选可去噪与缩小候选,也会排除小目标、遮挡严重对象或不同视角下的稀疏簇。模型匹配和图像信息能提供更多证据,但需要独立核对有效性,不能把“被保留”当已识别正确。

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

      Filtering the resulting clusters to identify objects of interest can be as simple as rejecting those that are too small (as shown in figure 14.1e), but it might also involve matching the points to a 3D model of a desired object or making use of image data (e.g., ICP and RANSAC from chapter 12).

    参考来源

    - 位置:《Introduction to Autonomous Robots Mechanisms, Sensors, Actuators, and Algorithms (Nikolaus Correll, Bradley Hayes etc.)》第14章《操纵》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Introduction to Autonomous Robots Mechanisms, Sensors, Actuators, and Algorithms (Nikolaus Correll, Bradley Hayes etc.)/_epub-src/OEBPS/xhtml/chapter_14.xhtml) - 结论依据:摘录中关于过滤条件可能删除真正目标的定义、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:Filtering the resulting clusters to identify objects of interest can be as simple as rejecting those that are too small (as shown in figure 14.1e), but it might also involve matching the points to a 3D model of a desired object or making use of image data (e.g., ICP and RANSAC from chapter 12).