PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation
Point cloud is an important type of geometric data structure. Due to its irregular format, most researchers transform such data to regular 3D voxel grids or collections of images. This, however, renders data unnecessarily voluminous and causes issues. In this paper, we design a novel type of neural network that directly consumes point clouds, which well respects the permutation invariance of points in the input. Our network, named PointNet, provides a unified architecture for applications ranging from object classification, part segmentation, to scene semantic parsing. Though simple, PointNet is highly efficient and effective. Empirically, it shows strong performance on par or even better than state of the art. Theoretically, we provide analysis towards understanding of what the network has learnt and why the network is robust with respect to input perturbation and corruption.
Видео PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation канала ComputerVisionFoundation Videos
Видео PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation канала ComputerVisionFoundation Videos
Показать
Комментарии отсутствуют
Информация о видео
12 августа 2017 г. 6:52:10
00:11:24
Другие видео канала
Tutorial : 3D Deep LearningSqueezeSeg demo: CNN for LiDAR point cloud segmentation3DmFV: 3D Point Cloud Classification in Real-Time using Convolutional Neural NetworksLecture 11 | Detection and SegmentationRangeNet++: Fast and Accurate LiDAR Semantic SegmentationDeep Learning for feature extraction from point clouds | AI & Cities | Dmitry KudinovOpen3D: A Modern Open-Source Library for 3D Data ProcessingDeep Models for 3D Reconstruction将门创投 | 斯坦福大学在读博士生祁芮中台:点云上的深度学习及其在三维场景理解中的应用Karl Iagnemma & Oscar Beijbom (Aptiv Autonomous Mobility) - MIT Self-Driving CarsWhat is a Point Cloud?Autoencoders - EXPLAINED3D Deep Learning TutorialUncertainty Quantification and Deep Learning ǀ Elise Jennings, Argonne National LaboratoryWhat's a Tensor?Point Voxel CNN for Efficient 3D Deep Learning, NeurIPS 2019, spotlight presentation.Point sampling - IPhilipp Krähenbühl - Point-based object detectionBuilding 3D deep learning models with PyTorch3D