1st Place Solutions for UG2+ Challenge 2022 ATMOSPHERIC TURBULENCE MITIGATION

TL;DR

This paper presents a unified end-to-end framework combining Restormer and NIMA modules for atmospheric turbulence mitigation, achieving top accuracy in the UG2+ Challenge 2022.

Contribution

The authors introduce a novel, efficient framework that effectively reconstructs high-quality images from turbulence-distorted frames, validated by extensive synthetic data.

Findings

Achieved 98.53% accuracy on text pattern reconstruction

Developed a synthetic dataset of over 10,000 images for training

Ranked 1st in the UG2+ Challenge 2022

Abstract

In this technical report, we briefly introduce the solution of our team ''summer'' for Atomospheric Turbulence Mitigation in UG$^2$+ Challenge in CVPR 2022. In this task, we propose a unified end-to-end framework to reconstruct a high quality image from distorted frames, which is mainly consists of a Restormer-based image reconstruction module and a NIMA-based image quality assessment module. Our framework is efficient and generic, which is adapted to both hot-air image and text pattern. Moreover, we elaborately synthesize more than 10 thousands of images to simulate atmospheric turbulence. And these images improve the robustness of the model. Finally, we achieve the average accuracy of 98.53\% on the reconstruction result of the text patterns, ranking 1st on the final leaderboard.