SOLVR: Submap Oriented LiDAR-Visual Re-Localisation

TL;DR

SOLVR is a unified LiDAR-Visual re-localisation pipeline that improves place recognition and registration accuracy across sensor modalities using a novel alignment, fusion, and registration strategy, achieving state-of-the-art results on KITTI datasets.

Contribution

It introduces a novel sensor alignment, a flexible training loss, and a robust registration method, advancing LiDAR-Visual re-localisation techniques.

Findings

Achieves state-of-the-art performance on KITTI datasets.

Improves registration accuracy over larger distances.

Enhances place recognition robustness across modalities.

Abstract

This paper proposes SOLVR, a unified pipeline for learning based LiDAR-Visual re-localisation which performs place recognition and 6-DoF registration across sensor modalities. We propose a strategy to align the input sensor modalities by leveraging stereo image streams to produce metric depth predictions with pose information, followed by fusing multiple scene views from a local window using a probabilistic occupancy framework to expand the limited field-of-view of the camera. Additionally, SOLVR adopts a flexible definition of what constitutes positive examples for different training losses, allowing us to simultaneously optimise place recognition and registration performance. Furthermore, we replace RANSAC with a registration function that weights a simple least-squares fitting with the estimated inlier likelihood of sparse keypoint correspondences, improving performance in scenarios with a low inlier ratio between the query and retrieved place. Our experiments on the KITTI and KITTI360 datasets show that SOLVR achieves state-of-the-art performance for LiDAR-Visual place recognition and registration, particularly improving registration accuracy over larger distances between the query and retrieved place.