GSTurb: Gaussian Splatting for Atmospheric Turbulence Mitigation

TL;DR

GSTurb is a novel framework that combines optical flow-guided tilt correction and Gaussian splatting to effectively mitigate atmospheric turbulence effects in long-range imaging, improving image quality in synthetic and real-world datasets.

Contribution

The paper introduces GSTurb, a new method integrating tilt correction and Gaussian splatting for turbulence mitigation, outperforming existing techniques in PSNR and SSIM metrics.

Findings

Achieves a peak PSNR of 27.67 dB and SSIM of 0.8735 on ATSyn-static dataset.

Improves PSNR by 1.3 dB and SSIM by 0.048 over the state-of-the-art.

Outperforms existing methods on real datasets like TSRWGAN and CLEAR.

Abstract

Atmospheric turbulence causes significant image degradation due to pixel displacement (tilt) and blur, particularly in long-range imaging applications. In this paper, we propose a novel framework for atmospheric turbulence mitigation, GSTurb, which integrates optical flow-guided tilt correction and Gaussian splatting for modeling non-isoplanatic blur. The framework employs Gaussian parameters to represent tilt and blur, and optimizes them across multiple frames to enhance restoration. Experimental results on the ATSyn-static dataset demonstrate the effectiveness of our method, achieving a peak PSNR of 27.67 dB and SSIM of 0.8735. Compared to the state-of-the-art method, GSTurb improves PSNR by 1.3 dB (a 4.5% increase) and SSIM by 0.048 (a 5.8% increase). Additionally, on real datasets, including the TSRWGAN Real-World and CLEAR datasets, GSTurb outperforms existing methods, showing significant improvements in both qualitative and quantitative performance. These results highlight that combining optical flow-guided tilt correction with Gaussian splatting effectively enhances image restoration under both synthetic and real-world turbulence conditions. The code for this method will be available at https://github.com/DuhlLiamz/3DGS_turbulence/tree/main.