知识卡片
光度一致需要静态与稳定外观
正确深度和相机运动也不能完整解释遮挡变化、移动对象、光照与反射变化。仅最小化像素差可能把这些变化错误吸收到深度或位姿。需要明确哪些像素适用几何重建假设,并为动态、可见性及外观变化采用额外处理。
这张卡片看不懂?
这张卡片已通过人工审查,不再接受失败反馈。
已有反馈 · 0 条
还没有人提交反馈。
在整书笔记中的位置
- · 单图深度估计
原书出处 · 1 处
- · 查看原书依据
If the reconstruction is correct, we expect the difference |ℓ1→2(p) – ℓ2(p)| to be very small. There will be some errors due to pixels that are invisible in one frame but not the other, changes in illumination, or due to the presence to moving objects in the scene.
参考来源
- 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第43章《单图深度估计》(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/c043.xhtml)
- 结论依据:摘录中关于光度一致需要静态与稳定外观的定义、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:If the reconstruction is correct, we expect the difference |ℓ1→2(p) – ℓ2(p)| to be very small. There will be some errors due to pixels that are invisible in one frame but not the other, changes in illumination, or due to the presence to moving objects in the scene.