Thermal Image Refinement with Depth Estimation using Recurrent Networks for Monocular ORB-SLAM3
This work addresses the problem of robust localization for UAVs in visually degraded conditions, offering an incremental improvement by integrating thermal refinement and depth estimation into an existing SLAM framework.
The paper tackles autonomous UAV navigation in GPS-denied, low-light environments by proposing a pipeline that uses a monocular thermal camera for depth estimation and SLAM, achieving competitive results such as an absolute relative error of approximately 0.06 on a radiometric dataset and a mean trajectory error under 0.4 m.
Autonomous navigation in GPS-denied and visually degraded environments remains challenging for unmanned aerial vehicles (UAVs). To this end, we investigate the use of a monocular thermal camera as a standalone sensor on a UAV platform for real-time depth estimation and simultaneous localization and mapping (SLAM). To extract depth information from thermal images, we propose a novel pipeline employing a lightweight supervised network with recurrent blocks (RBs) integrated to capture temporal dependencies, enabling more robust predictions. The network combines lightweight convolutional backbones with a thermal refinement network (T-RefNet) to refine raw thermal inputs and enhance feature visibility. The refined thermal images and predicted depth maps are integrated into ORB-SLAM3, enabling thermal-only localization. Unlike previous methods, the network is trained on a custom non-radiometric dataset, obviating the need for high-cost radiometric thermal cameras. Experimental results on datasets and UAV flights demonstrate competitive depth accuracy and robust SLAM performance under low-light conditions. On the radiometric VIVID++ (indoor-dark) dataset, our method achieves an absolute relative error of approximately 0.06, compared to baselines exceeding 0.11. In our non-radiometric indoor set, baseline errors remain above 0.24, whereas our approach remains below 0.10. Thermal-only ORB-SLAM3 maintains a mean trajectory error under 0.4 m.