TransLocNet: Cross-Modal Attention for Aerial-Ground Vehicle Localization with Contrastive Learning

TL;DR

TransLocNet introduces a cross-modal attention framework with contrastive learning to enhance aerial-ground vehicle localization, significantly reducing errors and achieving high accuracy across synthetic and real-world datasets.

Contribution

It presents a novel attention-based fusion method combined with contrastive learning for improved cross-modal localization between LiDAR and aerial imagery.

Findings

Reduces localization error by up to 63%

Achieves sub-meter and sub-degree accuracy

Outperforms state-of-the-art methods on CARLA and KITTI datasets

Abstract

Aerial-ground localization is difficult due to large viewpoint and modality gaps between ground-level LiDAR and overhead imagery. We propose TransLocNet, a cross-modal attention framework that fuses LiDAR geometry with aerial semantic context. LiDAR scans are projected into a bird's-eye-view representation and aligned with aerial features through bidirectional attention, followed by a likelihood map decoder that outputs spatial probability distributions over position and orientation. A contrastive learning module enforces a shared embedding space to improve cross-modal alignment. Experiments on CARLA and KITTI show that TransLocNet outperforms state-of-the-art baselines, reducing localization error by up to 63% and achieving sub-meter, sub-degree accuracy. These results demonstrate that TransLocNet provides robust and generalizable aerial-ground localization in both synthetic and real-world settings.