Neural-Pull: Learning Signed Distance Functions from Point Clouds by Learning to Pull Space onto Surfaces
<Neural-Pull>
Neural-pull: Learning signed distance functions from point clouds by learning to pull space onto surfaces目录
Motivation
- 训练一个神经网络去把query 3D locations “拉” 到他们在表面上的最近邻居；
拉的操作，方向是query locations处的网络梯度，步长是query locations处的网络SDF值，这两个都是从网络自身计算出来的 - 让我们可以同时更新sdf值和梯度
overview
- loss functions直接从GT点云本身定义，而不是利用GT SDF作回归；
