Abstract

Atmospheric turbulence severely degrades video quality by introducing distortions such as geometric warping, blur, and temporal flickering, posing significant challenges to both vi- sual clarity and temporal consistency. Current state-of-the-art methods are based on transformer, 3D architectures and re- quire multi-frame input, but their large computational cost and memory usage limit real-time deployment, especially in resource-constrained scenarios. In this work, we propose RMFAT — Recurrent Multi-scale Feature Atmospheric Tur- bulence Mitigator designed for efficient and temporally con- sistent video restoration under AT conditions. RMFAT adopts a lightweight recurrent framework that restores each frame using only two inputs at a time, significantly reducing tem- poral window size and computational burden. It further inte- grates multi-scale feature encoding and decoding with tem- poral warping modules at both encoder and decoder stages to enhance spatial detail and temporal coherence. Extensive ex- periments conducted on synthetic and real-world atmospheric turbulence datasets demonstrate that RMFAT not only outper- forms existing methods in terms of clarity restoration (with nearly a 9% improvement in SSIM) but also achieves sig- nificantly improved inference speed (achieving a more than fourfold reduction), making it particularly suitable for real- time atmospheric turbulence suppression tasks.

---

visual comparsion on real-world images

---

BibTeX

If you use our work, please cite:

    @misc{liu2025rmfatrecurrentmultiscalefeature,
          title={RMFAT: Recurrent Multi-scale Feature Atmospheric Turbulence Mitigator}, 
          author={Zhiming Liu and Nantheera Anantrasirichai},
          year={2025},
          eprint={2508.11409},
          archivePrefix={arXiv},
          primaryClass={cs.CV},
          url={https://arxiv.org/abs/2508.11409}, 
    }
  

[View on arXiv]