← 返回起始笔记位置

知识卡片

普通读书笔记卡 · 8908 · 生成模型:GPT-6

时间欠采样会反转运动

理解等级:中级

车轮在电影中看似倒转,来自离散时刻无法唯一分辨某些旋转速度与方向。时域抗混叠和提高采样率可以减少这类现象,分析时还需考虑物体的重复纹理。逐帧观察到的位移方向不一定等于真实连续运动方向。

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

      Spatiotemporal sampling can be studied with the tools we have already seen. In particular, equation (20.22) can explain sampling in the N-dimensional case. Temporal aliasing is responsible of the typical illusion in which we see wheels or fans changing the sense of rotation in movies. To avoid those artifacts it is important to apply an anti-aliasing filter as before.

    参考来源

    - 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第20章《图像采样与混叠》(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/c020b.xhtml) - 结论依据:摘录中关于时间欠采样会反转运动的定义、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:Spatiotemporal sampling can be studied with the tools we have already seen. In particular, equation (20.22) can explain sampling in the N-dimensional case. Temporal aliasing is responsible of the typical illusion in which we see wheels or fans changing the sense of rotation in movies. To avoid those artifacts it is important to apply an anti-aliasing filter as before.