• Corpus ID: 209324467

One Framework to Register Them All: PointNet Encoding for Point Cloud Alignment

@article{Sarode2019OneFT,
  title={One Framework to Register Them All: PointNet Encoding for Point Cloud Alignment},
  author={Vinit Sarode and Xueqian Li and Hunter Goforth and Yasuhiro Aoki and Animesh Dhagat and Rangaprasad Arun Srivatsan and Simon Lucey and Howie Choset},
  journal={ArXiv},
  year={2019},
  volume={abs/1912.05766}
}
PointNet has recently emerged as a popular representation for unstructured point cloud data, allowing application of deep learning to tasks such as object detection, segmentation and shape completion. However, recent works in literature have shown the sensitivity of the PointNet representation to pose misalignment. This paper presents a novel framework that uses PointNet encoding to align point clouds and perform registration for applications such as 3D reconstruction, tracking and pose… 
VRNet: Learning the Rectified Virtual Corresponding Points for 3D Point Cloud Registration
TLDR
A novel robust 3D point cloud registration framework that effectively breaks the distribution limitation of VCPs, and develops a hybrid loss function to enforce the shape and geometry structure consistency of the learned RCPs and the source to provide sufficient supervision.
A Representation Separation Perspective to Correspondence-Free Unsupervised 3-D Point Cloud Registration
TLDR
This letter proposes a correspondence-free unsupervised point cloud registration (UPCR) method that not only filters out the disturbance in pose-invariant representation but also is robust to partial-to-partial point clouds or noise.
ReAgent: Point Cloud Registration using Imitation and Reinforcement Learning
TLDR
This work proposes to consider iterative point cloud registration as a reinforcement learning task and presents a novel registration agent (ReAgent), which employs imitation learning to initialize its discrete registration policy based on a steady expert policy.
Pillar-based Object Detection for Autonomous Driving
TLDR
This work proposes a practical pillar-based approach to fix the imbalance issue caused by anchors, and incorporates a cylindrical projection into multi-view feature learning, predicts bounding box parameters per pillar rather than per point or per anchor, and includes an aligned pillar-to-point projection module to improve the final prediction.
Precise pose and assembly detection of generic tubular joints based on partial scan data
TLDR
A novel algorithm based on minimizing the area of a boundary enclosing partial scan data points is proposed for detecting both the pose and assembly of tubular joints with the aid of reference ideal models.
Vision-based Robotic Grasp Detection From Object Localization, Object Pose Estimation To Grasp Estimation: A Review.
TLDR
Three key tasks during robotic grasping are concluded, which are object localization, object pose estimation and grasp estimation, which contain object localization without classification, object detection and object instance segmentation.
Robust Partial-to-Partial Point Cloud Registration in a Full Range
TLDR
This work proposes GMCNet, which estimates pose-invariant correspondences for full-range Partial-to-Partial point cloud Registration (PPR) in the object-level registration scenario and proposes the Hierarchical Graphical Modeling architecture to encode robust descriptors, based on a synergy of hierarchical graph networks and graphical modeling.
End-to-end Learning the Partial Permutation Matrix for Robust 3D Point Cloud Registration
TLDR
A dedicated soft-to-hard (S2H) matching procedure within the registration pipeline consisting of two steps: solving the soft matching matrix and projecting this soft matrix to the partial permutation matrix (H-step), which creates a new state-of-the-art performance for robust 3D point cloud registration.
Vision-based robotic grasping from object localization, object pose estimation to grasp estimation for parallel grippers: a review
TLDR
Three key tasks during vision-based robotic grasping are concluded, which are object localization, object pose estimation and grasp estimation, which include 2D planar grasp methods and 6DoF grasp methods.

References

SHOWING 1-10 OF 46 REFERENCES
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
TLDR
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.
PointNetLK: Robust & Efficient Point Cloud Registration Using PointNet
TLDR
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.
3D Point Cloud Registration for Localization Using a Deep Neural Network Auto-Encoder
We present an algorithm for registration between a large-scale point cloud and a close-proximity scanned point cloud, providing a localization solution that is fully independent of prior information
Inverse Composition Discriminative Optimization for Point Cloud Registration
TLDR
This paper proposes Inverse Composition Discriminative Optimization (ICDO), an extension of Discrim inativeoptimization (DO), which learns a sequence of update steps from synthetic training data that search the parameter space for an improved solution.
VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detection
  • Yin Zhou, Oncel Tuzel
  • Computer Science, Environmental Science
    2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition
  • 2018
TLDR
VoxelNet is proposed, a generic 3D detection network that unifies feature extraction and bounding box prediction into a single stage, end-to-end trainable deep network and learns an effective discriminative representation of objects with various geometries, leading to encouraging results in3D detection of pedestrians and cyclists.
Iterative Transformer Network for 3D Point Cloud
TLDR
The Iterative Transformer Network (IT-Net) is proposed, a network module that canonicalizes the pose of a partial object with a series of 3D rigid transformations predicted in an iterative fashion and achieves superior performance over alternative 3D transformer networks on various tasks, such as partial shape classification and object part segmentation.
The Perfect Match: 3D Point Cloud Matching With Smoothed Densities
TLDR
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
TLDR
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.
PointNetVLAD: Deep Point Cloud Based Retrieval for Large-Scale Place Recognition
TLDR
This paper proposes a combination/modification of the existing PointNet and NetVLAD, which allows end-to-end training and inference to extract the global descriptor from a given 3D point cloud, and proposes the "lazy triplet and quadruplet" loss functions that can achieve more discriminative and generalizable global descriptors to tackle the retrieval task.
FlowNet 3 D : Learning Scene Flow in 3 D Point Clouds
TLDR
A novel deep neural network named FlowNet3D is proposed that learns scene flow from point clouds in an end-to-end fashion and successfully generalizes to real scans, outperforming various baselines and showing competitive results to the prior art.
...
1
2
3
4
5
...