Diffeomorphic Template Registration for Atmospheric Turbulence Mitigation

TL;DR

This paper introduces a robust, template-free diffeomorphic registration method for atmospheric turbulence mitigation, leveraging optical flow and flow inversion to improve image reconstruction without relying on supervised data or poor initial templates.

Contribution

The method models deformation via optical flow aggregation and introduces a flow inversion module to register images directly to a selected template without artifacts, achieving state-of-the-art results.

Findings

Decisive improvement in image registration quality

State-of-the-art reconstruction performance

Robustness with simple optical flow and reference selection

Abstract

We describe a method for recovering the irradiance underlying a collection of images corrupted by atmospheric turbulence. Since supervised data is often technically impossible to obtain, assumptions and biases have to be imposed to solve this inverse problem, and we choose to model them explicitly. Rather than initializing a latent irradiance ("template") by heuristics to estimate deformation, we select one of the images as a reference, and model the deformation in this image by the aggregation of the optical flow from it to other images, exploiting a prior imposed by Central Limit Theorem. Then with a novel flow inversion module, the model registers each image TO the template but WITHOUT the template, avoiding artifacts related to poor template initialization. To illustrate the robustness of the method, we simply (i) select the first frame as the reference and (ii) use the simplest optical flow to estimate the warpings, yet the improvement in registration is decisive in the final reconstruction, as we achieve state-of-the-art performance despite its simplicity. The method establishes a strong baseline that can be further improved by integrating it seamlessly into more sophisticated pipelines, or with domain-specific methods if so desired.