Abstract:Long-range imaging inevitably suffers from atmospheric turbulence with severe geometric distortions due to random refraction of light. The further the distance, the more severe the disturbance. Despite existing research has achieved great progress in tackling short-range turbulence, there is less attention paid to long-range turbulence with significant distortions. To address this dilemma and advance the field, we construct a large-scale real long-range atmospheric turbulence dataset (RLR-AT), including 1500 turbulence sequences spanning distances from 1 Km to 13 Km. The advantages of RLR-AT compared to existing ones: turbulence with longer-distances and higher-diversity, scenes with greater-variety and larger-scale. Moreover, most existing work adopts either registration-based or decomposition-based methods to address distortions through one-step mitigation. However, they fail to effectively handle long-range turbulence due to its significant pixel displacements. In this work, we propose a coarse-to-fine framework to handle severe distortions, which cooperates dynamic turbulence and static background priors (CDSP). On the one hand, we discover the pixel motion statistical prior of turbulence, and propose a frequency-aware reference frame for better large-scale distortion registration, greatly reducing the burden of refinement. On the other hand, we take advantage of the static prior of background, and propose a subspace-based low-rank tensor refinement model to eliminate the misalignments inevitably left by registration while well preserving details. The dynamic and static priors complement to each other, facilitating us to progressively mitigate long-range turbulence with severe distortions. Extensive experiments demonstrate that the proposed method outperforms SOTA methods on different datasets.

Geometric distortion correction of long-range imaging containing moving objects

A Fast Approach Against Large Foreground Motion in Real-time Image Stabilization

Unified Optical Distortion Correction Method for Imaging Systems Using A Concise Geometrical Transformation Model

Motion Blurred Image Restoration Based on Hartmann-Shack Wavefront Sensor

A New Approach for Real-Time Segmenting Moving Objects under Cluttered Background

Robust moving objects detection in long-distance imaging through turbulent medium

Restoration of TDI camera images with motion distortion and blur

Detecting and Tracking Dynamic Objects in Complex Environments

Detection and removal of lighting & shaking artifacts in home videos.

Long-range Turbulence Mitigation: A Large-scale Dataset and A Coarse-to-fine Framework

Dynamic turbulence mitigation for long-range imaging in the presence of large moving objects

Distortion-driven Turbulence Effect Removal using Variational Model.

Removing Turbulence Effect Via Hybrid Total Variation and Deformation-Guided Kernel Regression

Moving Object Extraction with a Hand-held Camera

Turb-Seg-Res: A Segment-then-Restore Pipeline for Dynamic Videos with Atmospheric Turbulence

A two-stage restoration of distorted underwater images using compressive sensing and image registration

A Fast Image Deformity Correction Algorithm for Underwater Turbulent Image Distortion

DeTurb: Atmospheric Turbulence Mitigation with Deformable 3D Convolutions and 3D Swin Transformers

Segmentation Technology of Ground Motion Target Based on Moving Camera

Imaging through the Atmosphere using Turbulence Mitigation Transformer

Underwater-to-air distorted image correction based on the reconstructed water surface