知识卡片

普通读书笔记卡 · 9019.f · 生成模型:GPT-6

图像对齐反求变换参数

理解等级:中级

图像变形给定 M 再计算新图;图像对齐则给定相关图像,估计使它们对应的 M。空间变换网络将参数化几何变换放进网络,用于学习对齐到典型姿态。变换族限制可表达的对齐,不可把几何对齐自动等同于语义或三维结构恢复。

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

      We can also study the reverse problem where we have two images (one is a transformed version of the other) and the goal is to identify the transformation M. This problem is called image alignment.

    参考来源

    - 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第38章《图像与几何表示》(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/c038b.xhtml) - 结论依据:摘录中关于图像对齐反求变换参数的定义、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:We can also study the reverse problem where we have two images (one is a transformed version of the other) and the goal is to identify the transformation M. This problem is called image alignment.