Atmospheric Turbulence Removal with Complex-Valued Convolutional Neural Network
This addresses the problem of real-time turbulence removal for dynamic scenes, which is incremental as it builds on deep learning approaches that were previously limited to static scenes.
The paper tackles atmospheric turbulence distortion in videos, which hinders visual interpretation, by proposing a novel learning-based framework using complex-valued convolutions to better capture phase information, and it significantly outperforms existing methods in mitigating distortions.
Atmospheric turbulence distorts visual imagery and is always problematic for information interpretation by both human and machine. Most well-developed approaches to remove atmospheric turbulence distortion are model-based. However, these methods require high computation and large memory making real-time operation infeasible. Deep learning-based approaches have hence gained more attention but currently work efficiently only on static scenes. This paper presents a novel learning-based framework offering short temporal spanning to support dynamic scenes. We exploit complex-valued convolutions as phase information, altered by atmospheric turbulence, is captured better than using ordinary real-valued convolutions. Two concatenated modules are proposed. The first module aims to remove geometric distortions and, if enough memory, the second module is applied to refine micro details of the videos. Experimental results show that our proposed framework efficiently mitigates the atmospheric turbulence distortion and significantly outperforms existing methods.