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

arXiv cs.RO / 2026/3/24

📰 ニュースSignals & Early TrendsIdeas & Deep AnalysisModels & Research

共有:

要点

The paper introduces TRGS-SLAM, an IMU-aided thermal SLAM system built on 3D Gaussian Splatting designed to work despite motion blur, rolling shutter distortions, and fixed pattern noise common to uncooled microbolometer thermal cameras.
It proposes a model-aware 3DGS rendering approach plus SLAM-specific innovations such as B-spline trajectory optimization with a two-stage IMU loss, view-diversity-based opacity resetting, and pose drift correction to improve robustness on degraded thermal imagery.
Experimental results claim accurate tracking in real-world fast-motion, high-noise thermal conditions where other tested SLAM methods fail.
The authors further report that offline refinement can restore thermal images with performance competitive to prior restoration methods that depended on ground-truth poses, indicating dual utility for both mapping and imaging quality improvement.

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.