知识卡片
滑窗用搜索覆盖位置尺度
滑窗在不同位置、尺度与长宽比裁剪候选,再用分类器判断。搜索网格越密覆盖更细,计算也增加;候选未覆盖的形状无法靠后续分类自动找回,旋转候选能扩展范围但有额外成本。
这张卡片看不懂?
这张卡片已通过人工审查,不再接受失败反馈。
已有反馈 · 0 条
还没有人提交反馈。
在整书笔记中的位置
- · 目标识别
原书出处 · 1 处
- · 查看原书依据
In its simplest form, the problem of object localization is posed as a binary classification task, namely distinguishing between a single object class and background class. Such a classification task can be turned into a detector by sliding it across the image (or image pyramid), and classifying each local window as shown in figure 50.11.
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:摘录中关于滑窗用搜索覆盖位置尺度的定义、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:In its simplest form, the problem of object localization is posed as a binary classification task, namely distinguishing between a single object class and background class. Such a classification task can be turned into a detector by sliding it across the image (or image pyramid), and classifying each local window as shown in figure 50.11.