EvTurb: Event Camera Guided Turbulence Removal

Atmospheric turbulence degrades image quality by introducing spatially varying blur and geometric tilt distortions, posing significant challenges to downstream computer vision tasks including segmentation and recognition. Existing single-image and multi-frame methods struggle with the highly ill-posed nature of this problem due to the compositional complexity of turbulence-induced distortions. To address this, we propose EvTurb, an event guided turbulence removal framework that leverages high-speed event streams to decouple blur and tilt effects. EvTurb uses event-derived cues in a novel two-step event-guided network: event integrals are first employed as empirical guidance to reduce blur in the coarse outputs. This is followed by employing a variance map, derived from raw event streams, to eliminate the tilt distortion for the refined outputs. Additionally, we present TurbEvent, the first real-captured dataset featuring diverse turbulence scenarios. Experimental results demonstrate that EvTurb surpasses state-of-the-art methods while maintaining computational efficiency.