← 返回起始笔记位置

知识卡片

普通读书笔记卡 · 9054.f · 生成模型:GPT-6

稀疏重建不是完整表面

理解等级:中级

稀疏SfM恢复可靠兴趣点及相机,为后续稠密处理提供几何基础,但点间表面、遮挡区域和纹理不足部分仍未恢复。把稀疏点云可视化得像物体不等于可直接导航或渲染完整表面,应用需求决定是否需要稠密、网格或其他表示。

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

      Multiview reconstruction is an active field of research with a wide range of applications in vision, graphics and healthcare. While a sparse SFM pipeline results in accurate 3D reconstruction, several applications (such as robotics and computer graphics) require dense 3D scene representations. The goal of dense SFM, not covered here, is to reconstruct every pixel from every image, as opposed to the sparse SFM systems

    参考来源

    - 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第44章《多视图几何与运动恢复结构》(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/c044.xhtml) - 结论依据:摘录中关于稀疏重建不是完整表面的定义、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:Multiview reconstruction is an active field of research with a wide range of applications in vision, graphics and healthcare. While a sparse SFM pipeline results in accurate 3D reconstruction, several applications (such as robotics and computer graphics) require dense 3D scene representations. The goal of dense SFM, not covered here, is to reconstruct every pixel from every image, as opposed to the sparse SFM systems