← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 8824.c · 生成模型:GPT-6

视觉表示可逐级显式化

理解等级:中级

Marr路线从图像强度进入边缘、交点等原始草图,再到可见深度与方向的二维半表示及三维对象结构。每阶段使不同信息显式化。深度学习可以学习中间表示,因此这条人工命名序列是历史理论,不是所有成功模型必须严格复制的层次。

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

      Primal sketch: Represents changes in the image (edges, junctions, zero-crossings) and their spatial organization. 2.5D sketch: Contains an estimation of the distance between the observer and the visible surfaces in the world at each pixel, also called a depth map. It also contains information about depth discontinuities and surface orientations. 3D model representation: Objects are represented as 3D shapes together with their spatial organization. You

    参考来源

    - 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第01章《视觉的挑战》(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/c001.xhtml) - 结论依据:摘录中关于视觉表示可逐级显式化的定义、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:Primal sketch: Represents changes in the image (edges, junctions, zero-crossings) and their spatial organization. 2.5D sketch: Contains an estimation of the distance between the observer and the visible surfaces in the world at each pixel, also called a depth map. It also contains information about depth discontinuities and surface orientations. 3D model representation: Objects are represented as 3D shapes together with their spatial organization. You