TRGS-SLAM: IMU-Aided Gaussian Splatting SLAM for Blurry, Rolling Shutter, and Noisy Thermal Images

TL;DR

TRGS-SLAM is a novel thermal SLAM system that effectively handles motion blur, rolling shutter effects, and noise in thermal images using Gaussian splatting and IMU data, enabling accurate mapping in challenging conditions.

Contribution

The paper introduces TRGS-SLAM, a thermal SLAM system with model-aware rendering and innovative optimization techniques, addressing thermal image degradations for the first time.

Findings

Achieves accurate tracking in high-noise, fast-motion thermal data.

Demonstrates thermal image restoration comparable to ground-truth-based methods.

Outperforms existing SLAM methods on challenging thermal datasets.

Abstract

Thermal cameras offer several advantages for simultaneous localization and mapping (SLAM) with mobile robots: they provide a passive, low-power solution to operating in darkness, are invariant to rapidly changing or high dynamic range illumination, and can see through fog, dust, and smoke. However, uncooled microbolometer thermal cameras, the only practical option in most robotics applications, suffer from significant motion blur, rolling shutter distortions, and fixed pattern noise. In this paper, we present TRGS-SLAM, a 3D Gaussian Splatting (3DGS) based thermal inertial SLAM system uniquely capable of handling these degradations. To overcome the challenges of thermal data, we introduce a model-aware 3DGS rendering method and several general innovations to 3DGS SLAM, including B-spline trajectory optimization with a two-stage IMU loss, view-diversity-based opacity resetting, and pose drift correction schemes. Our system demonstrates accurate tracking on real-world, fast motion, and high-noise thermal data that causes all other tested SLAM methods to fail. Moreover, through offline refinement of our SLAM results, we demonstrate thermal image restoration competitive with prior work that required ground truth poses.