Global visual localization in LiDAR-maps through shared 2D-3D embedding space

10/02/2019
by   Daniele Cattaneo, et al.
0

Global localization is an important and widely studied problem for many robotic applications. Place recognition approaches can be exploited to solve this task, e.g., in the autonomous driving field. While most vision-based approaches match an image w.r.t an image database, global visual localization within LiDAR-maps remains fairly unexplored, even though the path toward high definition 3D maps, produced mainly from LiDARs, is clear. In this work we leverage DNN approaches to create a shared embedding space between images and LiDAR-maps, allowing for image to 3D-LiDAR place recognition. We trained a 2D and a 3D Deep Neural Networks (DNNs) that create embeddings, respectively from images and from point clouds, that are close to each other whether they refer to the same place. An extensive experimental activity is presented to assess the effectiveness of the approach w.r.t. different learning methods, network architectures, and loss functions. All the evaluations have been performed using the Oxford Robotcar Dataset, which encompasses a wide range of weather and light conditions.

READ FULL TEXT
research
06/18/2021

Improved Radar Localization on Lidar Maps Using Shared Embedding

We present a heterogeneous localization framework for solving radar glob...
research
09/18/2023

RaLF: Flow-based Global and Metric Radar Localization in LiDAR Maps

Localization is paramount for autonomous robots. While camera and LiDAR-...
research
12/06/2022

Attention-Enhanced Cross-modal Localization Between 360 Images and Point Clouds

Visual localization plays an important role for intelligent robots and a...
research
06/27/2017

Topometric Localization with Deep Learning

Compared to LiDAR-based localization methods, which provide high accurac...
research
11/02/2022

A comparison of uncertainty estimation approaches for DNN-based camera localization

Camera localization, i.e., camera pose regression, represents a very imp...
research
05/29/2023

TReR: A Lightweight Transformer Re-Ranking Approach for 3D LiDAR Place Recognition

Autonomous driving systems often require reliable loop closure detection...
research
04/05/2018

Finding beans in burgers: Deep semantic-visual embedding with localization

Several works have proposed to learn a two-path neural network that maps...

Please sign up or login with your details

Forgot password? Click here to reset