MinkLoc3D: Point Cloud Based Large-Scale Place Recognition (2020-11-09T00:00:00.000000Z)

TL;DR

Evaluation on standard benchmarks proves that MinkLoc3D outperforms current state-of-the-art methods for computing a discriminative 3D point cloud descriptor, based on a sparse voxelized point cloud representation and sparse 3D convolutions.

Abstract

The paper presents a learning-based method for computing a discriminative 3D point cloud descriptor for place recognition purposes. Existing methods, such as Point-NetVLAD, are based on unordered point cloud representation. They use PointNet as the first processing step to extract local features, which are later aggregated into a global descriptor. The PointNet architecture is not well suited to capture local geometric structures. Thus, state-of-the-art methods enhance vanilla PointNet architecture by adding different mechanism to capture local contextual information, such as graph convolutional networks or using hand-crafted features. We present an alternative approach, dubbed Min-kLoc3D, to compute a discriminative 3D point cloud descriptor, based on a sparse voxelized point cloud representation and sparse 3D convolutions. The proposed method has a simple and efficient architecture. Evaluation on standard benchmarks proves that MinkLoc3D outperforms current state-of-the-art. Our code is publicly available on the project website. 1

Authors

J. Komorowski

2 papers

TL;DR

Abstract

Authors

References44 items

DH3D: Deep Hierarchical 3D Descriptors for Robust Large-Scale 6DoF Relocalization

DAGC: Employing Dual Attention and Graph Convolution for Point Cloud based Place Recognition

Deep Global Registration

A Metric Learning Reality Check

Revisiting Training Strategies and Generalization Performance in Deep Metric Learning

PyTorch: An Imperative Style, High-Performance Deep Learning Library

SuperGlue: Learning Feature Matching With Graph Neural Networks

Fully Convolutional Geometric Features

Deep Metric Learning to Rank

D2-Net: A Trainable CNN for Joint Description and Detection of Local Features

PCAN: 3D Attention Map Learning Using Contextual Information for Point Cloud Based Retrieval

4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks

Multi-Similarity Loss With General Pair Weighting for Deep Metric Learning

LPD-Net: 3D Point Cloud Learning for Large-Scale Place Recognition and Environment Analysis

From Coarse to Fine: Robust Hierarchical Localization at Large Scale

Symbolic Tensor Neural Networks for Digital Media – from Tensor Processing via BNF Graph Rules to CREAMS Applications

Dual Attention Network for Scene Segmentation

Benchmarking 6DOF Outdoor Visual Localization in Changing Conditions

PointNetVLAD: Deep Point Cloud Based Retrieval for Large-Scale Place Recognition

24/7 Place Recognition by View Synthesis

Dynamic Graph CNN for Learning on Point Clouds

Deep Metric Learning for Visual Understanding: An Overview of Recent Advances

Fine-Tuning CNN Image Retrieval with No Human Annotation

Squeeze-and-Excitation Networks

Random Erasing Data Augmentation

Sampling Matters in Deep Embedding Learning

Learnable pooling with Context Gating for video classification

PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space

Beyond Triplet Loss: A Deep Quadruplet Network for Person Re-identification

In Defense of the Triplet Loss for Person Re-Identification

1 year, 1000 km: The Oxford RobotCar dataset

Feature Pyramid Networks for Object Detection

PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation

NetVLAD: CNN Architecture for Weakly Supervised Place Recognition

VoxNet: A 3D Convolutional Neural Network for real-time object recognition

Sparse 3D convolutional neural networks

Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift

Shady dealings: Robust, long-term visual localisation using illumination invariance

Scalable 6-DOF Localization on Mobile Devices

Rank-based distance metric learning: An application to image retrieval

Signature Verification Using A "Siamese" Time Delay Neural Network

Flex-convolution

Signature veriﬁcation using a”

Batch normalization: Accelerating deep network training by reducing internal co-variate shift. volume 37 of Proceedings of Machine Learning Research , pages 448–456, Lille, France, 07–09 Jul 2015

Field of Study

Journal Information

Name

Page

Venue Information

Name

Type

URL

Alternate Names