ScoreBased Point Cloud Denoising (Learning Gradient Fields for Point Cloud Denoising)
Abstract
Point clouds acquired from scanning devices are often perturbed by noise, which affects downstream tasks such as surface reconstruction and analysis. The distribution of a noisy point cloud can be viewed as the distribution of a set of noisefree samples $p(x)$ convolved with some noise model $n$, leading to $(p * n)(x)$ whose mode is the underlying clean surface. To denoise a noisy point cloud, we propose to increase the loglikelihood of each point from $p * n$ via gradient ascent  iteratively updating each point's position. Since $p * n$ is unknown at testtime, and we only need the score (i.e., the gradient of the logprobability function) to perform gradient ascent, we propose a neural network architecture to estimate the score of $p * n$ given only noisy point clouds as input. We derive objective functions for training the network and develop a denoising algorithm leveraging on the estimated scores. Experiments demonstrate that the proposed model outperforms stateoftheart methods under a variety of noise models, and shows the potential to be applied in other tasks such as point cloud upsampling. The code is available at \url{https://github.com/luost26/scoredenoise}.
 Publication:

arXiv eprints
 Pub Date:
 July 2021
 arXiv:
 arXiv:2107.10981
 Bibcode:
 2021arXiv210710981L
 Keywords:

 Computer Science  Computer Vision and Pattern Recognition
 EPrint:
 ICCV 2021