Deep learning for 3D Point Cloud Analysis

Deep learning for 3D Point Cloud Analysis

Over the last few years, advances in graph, kernel, and sparse convolutions have helped establish deep networks as the predominant methods for 3D point clouds analysis.  In this talk, I first present the dynamic landscape of 3D deep learning, and introduce the superpoint graph approach for scaling memory-intensive algorithms to very large point clouds. Finally, I introduce the TorchPoints3D framework for easy and reproducible 3D deep learning, and present some applications of these approaches to remote sensing applications developed at IGN.