DeepSDF: Learning Continuous Signed Distance Functions for Shape Representation:学习用于形状表示的连续有符号距离函数 NeuS: Learning Neural Implicit Surfaces by Volume Rendering for Multi-view Reconstruction:通过体绘制学习神经隐式曲面用…
作者丨郭浩宇知乎 来源丨https://zhuanlan.zhihu.com/p/521725566 编辑丨3D视觉工坊 我们介绍一篇2022 CVPR Oral的三维场景重建论文:Neural 3D Scene Reconstruction with the Manhattan-world Assumption,该论文由浙江大学CAD&CG国家重点实验室/浙…