知识卡片
学习可补充模型未表达的线索
物体语义、单目深度与常见运动模式可帮助解释局部匹配不足的信息,学习方法能从数据中隐式利用这些线索。它同时可能学到训练分布偏差;“建筑通常静止”等经验不能作为任何输入下都正确的硬事实。
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
- · 学习运动估计
原书出处 · 1 处
- · 查看原书依据
Causes of modeling errors include failure of the brightness constancy assumption; the presence of occlusions, shadows and changes in illumination; new structures appearing due to changes in the resolution as a result of motion, deformable surfaces; and so on. Many of the motion computations involved approximations such as approximating the derivatives with finite size discrete convolutions. There could also be other motion-relevant cues present in
参考来源
- 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第49章《学习运动估计》(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/c049.xhtml)
- 结论依据:摘录中关于学习可补充模型未表达的线索的定义、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:Causes of modeling errors include failure of the brightness constancy assumption; the presence of occlusions, shadows and changes in illumination; new structures appearing due to changes in the resolution as a result of motion, deformable surfaces; and so on. Many of the motion computations involved approximations such as approximating the derivatives with finite size discrete convolutions. There could also be other motion-relevant cues present in