A unified Benchmark for Multi-Frame Image Restoration under Severe Refractive Warping

TL;DR

This paper introduces a comprehensive benchmark for evaluating video restoration methods under severe refractive distortions, including real and synthetic data, and assesses various algorithms with multiple metrics.

Contribution

It provides the first large-scale benchmark covering a wide range of refractive distortions, including new datasets, evaluation protocols, and analysis of diverse restoration methods.

Findings

Advanced learning-based methods outperform classical algorithms in severe distortions.

The benchmark reveals significant challenges in restoring highly refracted videos.

Perceptual metrics provide valuable insights alongside traditional pixel-based measures.

Abstract

Video sequence capturing through refractive dynamic media, such as a turbulent air or water surface, often suffer from severe geometric distortions and temporal instability. While recent advances address mild atmospheric turbulence, no existing benchmarks systematically evaluate restoration methods under strong and highly nonuniform refractive conditions. We present a comprehensive benchmark for geometric distortion removal in video, covering a range from turbulence-like mild warping to strong discontinuous refractive deformations. The benchmark includes both laboratory-captured real data and synthetic sequences generated for static scenes via physics-based light refraction modeling across four distortion levels and multiple surface wave types. We evaluate a spectrum of methods from simple baselines and classical registration algorithms to advanced learning-based approaches including DATUM and our proposed diffusion based V-cache for high and extreme distortions regimes. Evaluation uses both pixel-level (PSNR, SSIM), and perceptual (LPIPS, DINO, CLIP) metrics providing the first large scale analysis of geometric distortion removal. Our benchmark establishes a new foundation for developing and evaluating algorithms capable of reconstructing video from highly distorted optical environments. Our code and datasets are available at https://github.com/iafoss/refractive-mfir-benchmark.