Continuous Exposure-Time Modeling for Realistic Atmospheric Turbulence Synthesis

TL;DR

This paper introduces a novel continuous exposure-time model for atmospheric turbulence synthesis, creating a large-scale dataset that improves the realism and generalization of turbulence-affected image restoration models.

Contribution

It proposes a new exposure-time-dependent MTF and a comprehensive turbulence synthesis pipeline, resulting in the ET-Turb dataset for better turbulence modeling and restoration.

Findings

Models trained on ET-Turb outperform others in realism and generalization.

The ET-MTF accurately models blur as a continuous function of exposure-time.

The dataset enables improved turbulence effect simulation across diverse conditions.

Abstract

Atmospheric turbulence significantly degrades long-range imaging by introducing geometric warping and exposure-time-dependent blur, which adversely affects both visual quality and the performance of high-level vision tasks. Existing methods for synthesizing turbulence effects often oversimplify the relationship between blur and exposure-time, typically assuming fixed or binary exposure settings. This leads to unrealistic synthetic data and limited generalization capability of trained models. To address this gap, we revisit the modulation transfer function (MTF) formulation and propose a novel Exposure-Time-dependent MTF (ET-MTF) that models blur as a continuous function of exposure-time. For blur synthesis, we derive a tilt-invariant point spread function (PSF) from the ET-MTF, which, when integrated with a spatially varying blur-width field, provides a comprehensive and physically accurate characterization of turbulence-induced blur. Building on this synthesis pipeline, we construct ET-Turb, a large-scale synthetic turbulence dataset that explicitly incorporates continuous exposure-time modeling across diverse optical and atmospheric conditions. The dataset comprises 5,083 videos (2,005,835 frames), partitioned into 3,988 training and 1,095 test videos. Extensive experiments demonstrate that models trained on ET-Turb produce more realistic restorations and achieve superior generalization on real-world turbulence data compared to those trained on other datasets. The dataset is publicly available at: github.com/Jun-Wei-Zeng/ET-Turb.