Tranquil Clouds: Neural Networks for Learning Temporally Coherent Features in Point Clouds

Summary

Permutation invariant loss terms for temporally coherent point set generation
A Siamese training setup and generator architecture for point-based super-resolution with neural networks
Enabling improved output variance by allowing for dynamic adjustments of the output size

Abstract

Point clouds, as a form of Lagrangian representation, allow for powerful and flexible applications in a large number of computational disciplines. We propose a novel deep-learning method to learn stable and temporally coherent feature spaces for points clouds that change over time. We identify a set of inherent problems with these approaches: without knowledge of the time dimension, the inferred solutions can exhibit strong flickering, and easy solutions to suppress this flickering can result in undesirable local minima that manifest themselves as halo structures. We propose a novel temporal loss function that takes into account higher time derivatives of the point positions, and encourages mingling, i.e., to prevent the aforementioned halos. We combine these techniques in a super-resolution method with a truncation approach to flexibly adapt the size of the generated positions. We show that our method works for large, deforming point sets from different sources to demonstrate the flexibility of our approach.

Author

Journal/Conference

Subjects

cs.CV: Computer Vision and Pattern Recognition
cs.GR: Graphics
cs.LG: Machine Learning
stat.ML: Machine Learning

Comment

Further information and videos at this https URL

Link

arXiv: https://arxiv.org/abs/1907.05279
Project: https://ge.in.tum.de/publications/2020-iclr-prantl/
Code: https://gitlab.com/Prantl/NeuralParticles

j20232 / survey