知识卡片
共识筛选后再精化模型
两点随机候选只作初步假设,可用其内点重新进行最小二乘或其他优化来提高精度。随后还可重新核验内点,因为模型更新会改变残差;初始随机采样与最终估计各有职责,不能用一个随机点对作为最终精确答案。
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
- · 特征提取
原书出处 · 1 处
- · 查看原书依据
Given that RANSAC is random, finding a really good fit can be computationally intensive and time-consuming. Therefore, RANSAC is usually used only as a first step to get an initial estimate, which can then be improved by some kind of local optimization, such as least squares.
参考来源
- 位置:《Introduction to Autonomous Robots Mechanisms, Sensors, Actuators, and Algorithms (Nikolaus Correll, Bradley Hayes etc.)》第09章《特征提取》(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_9.xhtml)
- 结论依据:摘录中关于共识筛选后再精化模型的定义、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:Given that RANSAC is random, finding a really good fit can be computationally intensive and time-consuming. Therefore, RANSAC is usually used only as a first step to get an initial estimate, which can then be improved by some kind of local optimization, such as least squares.