pcdiff 1.0.0

Simple and small library to compute differential operators (gradient, divergence, Laplacian) on point clouds.

pcdiff: Differential operators on point clouds

Simple and small library to compute differential operators (gradient, divergence, Laplacian) on point clouds.

Installation

The recommended installation method is by using pip:

pip install ./

Example usage

Import the required functionality and get your gradient and divergence matrices:

import numpy as np
from pcdiff import knn_graph, estimate_basis, build_grad_div

# Random point cloud
pos = np.random.rand(1000, 3)

# Generate kNN graph
edge_index = knn_graph(pos, 20)
# Estimate normals and local frames
basis = estimate_basis(pos, edge_index)
# Build gradient and divergence operators (Scipy sparse matrices)
grad, div = build_grad_div(pos, *basis, edge_index)

# ... use gradient and divergence in any task you like

For sake of simplicity, every operation is written in Numpy and can be accelerated with Numba or Jax. If you would like to use these operators in PyTorch, please refer the github repository for DeltaConv: pip install deltaconv and use the operators from deltaconv.geometry.

Citation

If you find this library useful in your own work, please cite our paper on DeltaConv, a convolution for point clouds that uses these operators:

@Article{Wiersma2022DeltaConv,
  author    = {Ruben Wiersma, Ahmad Nasikun, Elmar Eisemann, Klaus Hildebrandt},
  journal   = {Transactions on Graphics},
  title     = {DeltaConv: Anisotropic Operators for Geometric Deep Learning on Point Clouds},
  year      = {2022},
  month     = jul,
  number    = {4},
  volume    = {41},
  doi       = {10.1145/3528223.3530166},
  publisher = {ACM},
}