BriefGPT.xyz
Sep, 2021
CodeNeRF: 面向物体类别的可分离神经辐射场
CodeNeRF: Disentangled Neural Radiance Fields for Object Categories
HTML
PDF
Wonbong Jang, Lourdes Agapito
TL;DR
CodeNeRF利用3D神经表征来学习物体的形状和纹理变化，从而可以合成未知物体的新视角，并且可以通过优化来估计摄像机视角、形状和外观代码。
Abstract
codenerf
is an
implicit 3d neural representation
that learns the variation of object shapes and textures across a category and can be trained, from a set of posed images, to synthesize novel views of unseen objec
→