Survey

Advances in Neural Rendering [link]

Neural Fields in Visual Computing and Beyond [link]

SDF的不同替代

Neural unsigned distance fields for implicit function learning [NIPS20]

Neural-Pull: Learning Signed Distance Functions from Point Clouds by Learning to Pull Space onto Surfaces [ICML21]

学query点到点云最近点的距离

Deep Implicit Surface Point Prediction Networks [ICCV21]

SDF，UDF的替代，学习输入点在surface上的最近点

Representing 3D Shapes with Probabilistic Directed Distance Fields [CVPR22]

给定点和方向，直接预测可见性和到surface的距离，没给代码

Learning Consistency-Aware Unsigned Distance Functions Progressively from Raw Point Clouds [NIPS22]

Neural-Pull改了个loss

3PSDF: Three-Pole Signed Distance Function for Learning Surfaces with Arbitrary Topologies [CVPR22]

八叉树结构，没有surface的体素距离为NAN，有的则在体素内部用SDF表示

GIFS: Neural Implicit Function for General Shape Representation [CVPR22]

用空间中两个query点之间是否穿过surface来表达，怀疑效果

#

Differentiable Point-Based Radiance Fields for Efficient View Synthesis [SIGGRAPHaisa22]

ADOP: Approximate Differentiable One-Pixel Point Rendering [TOG22]

#

A Level Set Theory for Neural Implicit Evolution under Explicit Flows [ECCV22]

ParticleNeRF: Particle Based Encoding for Online Neural Radiance Fields in Dynamic Scenes

用带有特征的运动粒子来做动态场景NeRF

Differentiable Surface Rendering via Non-Differentiable Sampling [ICCV21]

a method for differentiable rendering of 3D surfaces that supports both explicit and implicit representations.

Recovering Fine Details for Neural Implicit Surface Reconstruction [link]