Cross-Modal Visual Relocalization in Prior LiDAR Maps Utilizing Intensity Textures

TL;DR

This paper introduces a cross-modal visual relocalization system that leverages intensity textures from prior LiDAR maps, improving place recognition and pose estimation by addressing texture-geometry inconsistencies.

Contribution

It proposes a novel system utilizing intensity textures in LiDAR maps with a three-module approach for improved relocalization accuracy.

Findings

Effective in place recognition and pose estimation

Outperforms existing methods on self-collected datasets

Robust 6DoF pose estimation achieved

Abstract

Cross-modal localization has drawn increasing attention in recent years, while the visual relocalization in prior LiDAR maps is less studied. Related methods usually suffer from inconsistency between the 2D texture and 3D geometry, neglecting the intensity features in the LiDAR point cloud. In this paper, we propose a cross-modal visual relocalization system in prior LiDAR maps utilizing intensity textures, which consists of three main modules: map projection, coarse retrieval, and fine relocalization. In the map projection module, we construct the database of intensity channel map images leveraging the dense characteristic of panoramic projection. The coarse retrieval module retrieves the top-K most similar map images to the query image from the database, and retains the top-K' results by covisibility clustering. The fine relocalization module applies a two-stage 2D-3D association and a covisibility inlier selection method to obtain robust correspondences for 6DoF pose estimation. The experimental results on our self-collected datasets demonstrate the effectiveness in both place recognition and pose estimation tasks.