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

2026-03-20

AI Summary

This work proposes a thermal-inertial SLAM system based on 3D Gaussian Splatting (3DGS) to address the failure of localization and mapping caused by motion blur, rolling shutter distortion, and fixed-pattern noise in uncooled microbolometer imagery. The approach integrates model-aware thermal rendering, B-spline continuous trajectory modeling, IMU preintegration constraints, and explicit rolling shutter modeling, complemented by a viewpoint-diversity-driven opacity reset and pose drift correction mechanism. Experimental results demonstrate that the system achieves robust and accurate tracking under high-speed motion and high-noise thermal imaging conditions, outperforming existing SLAM methods. Furthermore, its offline optimization yields reconstruction quality comparable to thermal image restoration techniques that rely on ground-truth poses.

Technology Category

• Computer Vision: Motion & Tracking
• Intelligent Robots: Localization, Mapping, and Navigation
• Machine Learning: Calibration & Uncertainty Quantification

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Security and Privacy: Tracking, profiling, and countermeasures against them

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.

Problem

• thermal SLAM
• motion blur
• rolling shutter
• fixed pattern noise
• 3D Gaussian Splatting

Innovation

• 3D Gaussian Splatting
• Thermal SLAM
• IMU-Aided Optimization
• Rolling Shutter Correction
• Motion Blur Robustness