ThermalLoc: A Vision Transformer-Based Approach for Robust Thermal Camera Relocalization in Large-Scale Environments

TL;DR

ThermalLoc is a new deep learning approach that combines EfficientNet and Transformers to improve thermal camera relocalization accuracy and robustness in large-scale environments.

Contribution

It introduces ThermalLoc, the first end-to-end deep learning method specifically designed for thermal image relocalization, integrating local and global feature extraction.

Findings

ThermalLoc outperforms existing methods like AtLoc, MapNet, PoseNet, and RobustLoc.

It achieves higher accuracy and robustness in thermal camera relocalization.

Evaluations on multiple datasets validate its effectiveness.

Abstract

Thermal cameras capture environmental data through heat emission, a fundamentally different mechanism compared to visible light cameras, which rely on pinhole imaging. As a result, traditional visual relocalization methods designed for visible light images are not directly applicable to thermal images. Despite significant advancements in deep learning for camera relocalization, approaches specifically tailored for thermal camera-based relocalization remain underexplored. To address this gap, we introduce ThermalLoc, a novel end-to-end deep learning method for thermal image relocalization. ThermalLoc effectively extracts both local and global features from thermal images by integrating EfficientNet with Transformers, and performs absolute pose regression using two MLP networks. We evaluated ThermalLoc on both the publicly available thermal-odometry dataset and our own dataset. The results demonstrate that ThermalLoc outperforms existing representative methods employed for thermal camera relocalization, including AtLoc, MapNet, PoseNet, and RobustLoc, achieving superior accuracy and robustness.