知识卡片
无训练算法仍需成像假设
梯度、光度配准与分层方法不必先用数据训练网络,但依赖亮度恒常、小残余运动、局部一致性及正则等设计。它们的重建目标也可作为学习方法的监督信号;没有人工标签不等于没有假设或无需验收真实运动。
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
- · 光流估计
原书出处 · 1 处
- · 查看原书依据
The approaches presented here require no training. In the next chapter, we will study several learning-based methods for motion estimation. The approaches presented in this chapter will become useful when exploring unsupervised learning methods.
参考来源
- 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第48章《光流估计》(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/c048.xhtml)
- 结论依据:摘录中关于无训练算法仍需成像假设的定义、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:The approaches presented here require no training. In the next chapter, we will study several learning-based methods for motion estimation. The approaches presented in this chapter will become useful when exploring unsupervised learning methods.