3D Point Cloud Registration with Multi-Scale Architecture and Unsupervised Transfer Learning

  title={3D Point Cloud Registration with Multi-Scale Architecture and Unsupervised Transfer Learning},
  author={Sofiane Horache and Jean-Emmanuel Deschaud and Franccois Goulette},
  journal={2021 International Conference on 3D Vision (3DV)},
We propose a method for generalizing deep learning for 3D point cloud registration on new, totally different datasets. It is based on two components, MS-SVConv and UDGE. Using Multi-Scale Sparse Voxel Convolution, MSSVConv is a fast deep neural network that outputs the descriptors from point clouds for 3D registration between two scenes. UDGE is an algorithm for transferring deep networks on unknown datasets in a unsupervised way. The interest of the proposed method appears while using the two… 
Centimeter-level Positioning by Instantaneous Lidar-aided GNSS Ambiguity Resolution
—High-precision vehicle positioning is key to the implementation of modern driving systems in urban environments. Global Navigation Satellite System (GNSS) carrier phase measurements can provide


Self-Supervised Point Set Local Descriptors for Point Cloud Registration
The capability of the self-supervised local descriptor to achieve even better performance than the supervised model, while being easier to train and requiring no data labeling is demonstrated.
The Perfect Match: 3D Point Cloud Matching With Smoothed Densities
This work proposes 3DSmoothNet, a full workflow to match 3D point clouds with a siamese deep learning architecture and fully convolutional layers using a voxelized smoothed density value (SDV) representation, and shows that 3DS moothNet trained only on RGB-D indoor scenes of buildings achieves 79.0% average recall, more than double the performance of the closest, learning-based competitors.
PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation
This paper designs a novel type of neural network that directly consumes point clouds, which well respects the permutation invariance of points in the input and provides a unified architecture for applications ranging from object classification, part segmentation, to scene semantic parsing.
Deep Closest Point: Learning Representations for Point Cloud Registration
  • Yue Wang, J. Solomon
  • Computer Science
    2019 IEEE/CVF International Conference on Computer Vision (ICCV)
  • 2019
This work proposes a learning-based method, titled Deep Closest Point (DCP), inspired by recent techniques in computer vision and natural language processing, that provides a state-of-the-art registration technique and evaluates the suitability of the learned features transferred to unseen objects.
End-to-End Learning Local Multi-View Descriptors for 3D Point Clouds
This work proposes an end-to-end framework to learn local multi-view descriptors for 3D point clouds by using a differentiable renderer, which allows the viewpoints to be optimizable parameters for capturing more informative local context of interest points.
D3Feat: Joint Learning of Dense Detection and Description of 3D Local Features
This paper proposes a keypoint selection strategy that overcomes the inherent density variations of 3D point clouds, and proposes a self-supervised detector loss guided by the on-the-fly feature matching results during training.
Self-Supervised Pretraining of 3D Features on any Point-Cloud
This work presents a simple self-supervised pretraining method that can work with single-view depth scans acquired by varied sensors, without 3D registration and point correspondences, and sets a new state-of-the-art for object detection on ScanNet and SUNRGBD.
DeepVCP: An End-to-End Deep Neural Network for Point Cloud Registration
DeepVCP is presented - a novel end-to-end learning-based 3D point cloud registration framework that achieves comparable registration accuracy to prior state-of-the-art geometric methods and the low registration error and high robustness of the method make it attractive to the substantial applications relying on the point cloudRegistration task.
Classification of Point Cloud Scenes with Multiscale Voxel Deep Network
A new convolutional neural network (CNN) to classify 3D point clouds of urban or indoor scenes that beats the state of the art of point classification methods (those not using a regularization step).
PointNetLK: Robust & Efficient Point Cloud Registration Using PointNet
It is argued that PointNet itself can be thought of as a learnable "imaging" function, and classical vision algorithms for image alignment can be brought to bear on the problem -- namely the Lucas & Kanade (LK) algorithm.