# Topometric Localization with Deep Learning

## TL;DR

This paper introduces a deep learning-based visual localization method that leverages LiDAR data for training, achieving high accuracy comparable to LiDAR-based systems while using only cameras, and demonstrates significant error reduction in challenging conditions.

## Contribution

The paper presents a novel vision-based localization approach trained on LiDAR outputs, combining deep networks for visual odometry and topological localization with optimization.

## Key findings

- Localization errors up to 10 times smaller than traditional vision methods
- Effective in varying weather conditions over six months
- Utilizes a new challenging pedestrian dataset

## Abstract

Compared to LiDAR-based localization methods, which provide high accuracy but rely on expensive sensors, visual localization approaches only require a camera and thus are more cost-effective while their accuracy and reliability typically is inferior to LiDAR-based methods. In this work, we propose a vision-based localization approach that learns from LiDAR-based localization methods by using their output as training data, thus combining a cheap, passive sensor with an accuracy that is on-par with LiDAR-based localization. The approach consists of two deep networks trained on visual odometry and topological localization, respectively, and a successive optimization to combine the predictions of these two networks. We evaluate the approach on a new challenging pedestrian-based dataset captured over the course of six months in varying weather conditions with a high degree of noise. The experiments demonstrate that the localization errors are up to 10 times smaller than with traditional vision-based localization methods.