Nerf October 14, 2021 less than 1 minute read NerfPermalink Repository spark star Nerf HomePage Paper.pdf Key PointPermalink 输入是位置和视角方向(x,y,z)+(θ,ϕ)(x,y,z) + (\theta, \phi)(x,y,z)+(θ,ϕ)构成的5D坐标，输出是体积密度和颜色 Position encoding可以提高细节 层次化的采样流程可以减少采样 通过volume rendering实现视图合成 全连接网络MLP ReferencePermalink 【论文精炼】何雨龙 - cnblogs nerf(ML Research Paper Explained) - Youtube 图形学新高潮? NeRF 笔记 - Lee bro的文章 - 知乎 NeRF ——使用神经辐射场(Neural Radiance Fields)表示场景来进行视图合成(View Synthesis)🤙 【NeRF论文笔记】用于视图合成的神经辐射场技术 - 林天威的文章 - 知乎 Twitter Facebook LinkedIn Previous Next
FruitNerf September 28, 2024 2 minute read FruitNeRF: A Unified Neural Radiance Field based Fruit Counting Framework
Unsigned Orthogonal Distance Fields: An Accurate Neural Implicit Representation for Diverse 3D Shapes May 13, 2024 less than 1 minute read
