知识卡片

普通读书笔记卡 · 9032.b · 生成模型:GPT-6

极点由左右奇异向量求取

理解等级:高级

若F=UΣV^T,第一图极点满足Fe1=0,取最小奇异值对应右奇异向量;第二图极点满足F^T e2=0,取对应左奇异向量。E同理但结果为归一化坐标。噪声下应求奇异向量,不是原文所说直接取E、E^T最小特征值向量。

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

      As the matrix E is likely to be noisy, we can compute the epipoles as the eigenvectors with the smallest eigenvalues of E and ET. 40.3.4.4 Epipolar lines: The game In order to build an intuition of how epipolar lines work we propose to play a game. Figure 40.13 shows a set of camera pairs and a set of epipolar lines. Can you guess what

    参考来源

    - 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第40章《立体视觉》(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/c040.xhtml) - 结论依据:摘录中关于极点由左右奇异向量求取的定义、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:As the matrix E is likely to be noisy, we can compute the epipoles as the eigenvectors with the smallest eigenvalues of E and ET. 40.3.4.4 Epipolar lines: The game In order to build an intuition of how epipolar lines work we propose to play a game. Figure 40.13 shows a set of camera pairs and a set of epipolar lines. Can you guess what