FIReStereo: Forest InfraRed Stereo Dataset for UAS Depth Perception in Visually Degraded Environments

Abstract

Robust depth perception in visually-degraded environments is crucial for autonomous aerial systems. Thermal imaging cameras, which capture infrared radiation, are robust to visual degradation. However, due to lack of a large-scale dataset, the use of thermal cameras for unmanned aerial system (UAS) depth perception has remained largely unexplored. This paper presents a stereo thermal depth perception dataset for autonomous aerial perception applications. The dataset consists of stereo thermal images, LiDAR, IMU and ground truth depth maps captured in urban and forest settings under diverse conditions like day, night, rain, and smoke. We benchmark representative stereo depth estimation algorithms, offering insights into their performance in degraded conditions. Models trained on our dataset generalize well to unseen smoky conditions, highlighting the robustness of stereo thermal imaging for depth perception. We aim for this work to enhance robotic perception in disaster scenarios, allowing for exploration and operations in previously unreachable areas. The dataset and source code are available at https://firestereo.github.io.

Figures (9)

• Figure 1: Overview of the UAS collection setup, various outdoor environments, and sensor modalities in the FIReStereo Dataset.
• Figure 2: Comparison of sensors under smoke. Visual and LiDAR sensors are severely affected by particulates, while long-wave infrared is able to bypass the infrared signature from smoke, yielding clear visibility of the environment.
• Figure 3: Data collection setup with stereo thermal pair, LiDAR, IMU, and AGX Orin.
• Figure 4: Maps generated using LiDAR-SLAM, which is used to obtain ground truth depth in \ref{['fig:data_variety']}.
• Figure 5: Photos of different environments from our dataset featuring different tree densities, light conditions, and smoke conditions.