← 返回起始笔记位置

知识卡片

普通读书笔记卡 · 9079.c · 生成模型:GPT-6

直接预测也包含架构先验

理解等级:中级

FlowNet式端到端网络直接从两帧回归流,省去手工指定每个匹配步骤,却仍受卷积、瓶颈、感受野与损失等设计约束。“不预设所需运动模块”不等于完全没有任何架构归纳偏置。

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

      The direct approach depicted in figure 49.1 learns to estimate optical flow directly from a pair of frames. This architecture makes no assumptions about which architectural priors are needed to compute optical flow from images. The architecture is trained end-to-end using ground truth optical flow.

    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:摘录中关于直接预测也包含架构先验的定义、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 direct approach depicted in figure 49.1 learns to estimate optical flow directly from a pair of frames. This architecture makes no assumptions about which architectural priors are needed to compute optical flow from images. The architecture is trained end-to-end using ground truth optical flow.