连续曝光时间建模用于真实大气湍流合成
Continuous Exposure-Time Modeling for Realistic Atmospheric Turbulence Synthesis
提出曝光时间相关的调制传递函数（ET-MTF）和倾斜不变的点扩散函数（PSF），用于生成物理精确的大气湍流模糊，可复用于图像复原任务的合成数据生成与泛化性评估。
原文依据 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.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论大气湍流通过引入几何扭曲和曝光时间相关的模糊，显著降低远距离成像质量，对视觉质量和高级视觉任务性能均产生不利影响。现有的大气湍流效应合成方法通常过度简化模糊与曝光时间的关系，往往假设固定或二值曝光设置。这导致合成数据不真实，并限制了训练模型的泛化能力。为弥补这一不足，我们重新审视调制传递函数（MTF）公式，提出一种新颖的曝光时间相关MTF（ET-MTF），将模糊建模为曝光时间的连续函数。在模糊合成方面，我们从ET-MTF推导出倾斜不变的点扩散函数（PSF），并与空间变化的模糊宽度场结合，提供了对湍流引起的模糊的全面且物理准确的表征。基于该合成流程，我们构建了ET-Turb，一个大规模合成湍流数据集，明确纳入了跨多种光学和大气条件的连续曝光时间建模。该数据集包含5,083个视频（2,005,835帧），划分为3,988个训练视频和1,095个测试视频。大量实验表明，与在其他数据集上训练的模型相比，在ET-Turb上训练的模型能产生更真实的复原结果，并在真实世界湍流数据上取得更优的泛化性能。数据集已公开：github.com/Jun-Wei-Zeng/ET-Turb。
英文原文摘要 1557 字符
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.