← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 9051.b · 生成模型:GPT-6

同一设备未必共享有效内参

理解等级:中级

同相机固定成像模式时共享K可减少未知量,但变焦、对焦相关变化、裁剪、图像缩放或不同镜头会改变有效内参。应按实际模式和画幅分组,而不是单凭设备名称强制所有图像用相同K。

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

      For each of the M views (it could be different cameras or different pictures taking by a moving camera), we denote the projection of point Pi as seen by camera j by . Each view has camera parameters K(j), R(j), and T(j). If all the images are captured by the same camera, then the matrices K(j) will be the same. In general it is not

    参考来源

    - 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第44章《多视图几何与运动恢复结构》(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/c044.xhtml) - 结论依据:摘录中关于同一设备未必共享有效内参的定义、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:For each of the M views (it could be different cameras or different pictures taking by a moving camera), we denote the projection of point Pi as seen by camera j by . Each view has camera parameters K(j), R(j), and T(j). If all the images are captured by the same camera, then the matrices K(j) will be the same. In general it is not