← 返回起始笔记位置

知识卡片

普通读书笔记卡 · 8966.c · 生成模型:GPT-6

离散视差精度受分箱限制

理解等级:中级

书中将视差粗分为四个宽区间,再对区间内局部证据汇总,便于展示消息和边缘分布。输出只能反映该离散状态空间,不应当作精确连续深度。比较结果时还需明确视差到深度的几何映射及区间所代表的值。

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 simplicity in this example, we discretize the disparity offsets into four different bins, each 45 disparity pixels wide. We sum the local disparity evidence within each depth bin, then normalize the evidence at each position to maintain a probability distribution. The resulting local evidence matrix, for each of the four depth states, at each left camera spatial position, is displayed in figure 29.21(c). The

    参考来源

    - 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第29章《概率图模型》(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/c029.xhtml) - 结论依据:摘录中关于离散视差精度受分箱限制的定义、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 simplicity in this example, we discretize the disparity offsets into four different bins, each 45 disparity pixels wide. We sum the local disparity evidence within each depth bin, then normalize the evidence at each position to maintain a probability distribution. The resulting local evidence matrix, for each of the four depth states, at each left camera spatial position, is displayed in figure 29.21(c). The