基于因果的图像退化鲁棒性评估分类学
A Causally Grounded Taxonomy for Image Degradation Robustness Evaluation
AI总结：
本文提出一个基于因果的图像退化分类学，通过双轴抽象整合算法退化、感知失真和物理退化，提供统一的严重性度量层以提升跨数据集和任务的可比性。
摘要：
图像退化可能在采集、处理和传输过程中发生，改变视觉外观并影响下游视觉任务。这些退化现象在多个领域被研究，包括合成污染基准测试、感知图像质量评估以及物理基础的成像系统或真实相机故障分析。尽管这些领域处理密切相关现象，但常使用不兼容的分组方案和后端特定严重性定义，导致结果难以跨数据集、退化源和任务比较。本文提出一个基于因果的框架，用于组织和解释这些设置中的图像退化。我们提供一个解释性表示和度量层，使隐含假设显性化。每个退化沿两个正交轴描述：成像管道中的主导因果源（环境、传感器/光学、ISP/渲染器/编码器或传输/系统）及其感知效果。这种双轴抽象产生了一个紧凑的分类学，涵盖算法退化、感知失真和物理驱动的成像伪影。为解决不一致的严重性语义而不改变现有实现，我们引入了一个轻量级的严重性度量层。对于每个退化和每个给定后端的原生严重性级别，我们使用全参考图像质量指标（PSNR、SSIM和LPIPS）量化退化强度。这使严重性在不同源之间可观察和可比，同时保留原生参数化。我们通过COCO Degradation展示该框架，这是一个对齐基准，用于评估在多样成像条件下目标检测器的鲁棒性。
英文摘要：
Image degradations can occur during acquisition, processing, and transmission, altering visual appearance and affecting downstream vision tasks. They are studied in several communities, including synthetic corruption benchmarks for robustness evaluation, perceptual image quality assessment, and physically grounded analyses of imaging systems or real camera failures. Although these areas address closely related phenomena, they often use incompatible grouping schemes and backend specific severity definitions, making results difficult to compare across datasets, degradation sources, and tasks. We propose a causally grounded framework for organizing and interpreting image degradations across these settings. Instead of introducing new degradations or redefining existing benchmarks, we provide an interpretive representation and measurement layer that makes implicit assumptions explicit. Each degradation is described along two orthogonal axes: its dominant causal source in the imaging pipeline (environment, sensor/optics, ISP/renderer/codec, or transfer/system), and its resulting perceptual effect. This dual axis abstraction yields a compact taxonomy spanning algorithmic corruptions, perceptual distortions, and physically motivated imaging artifacts. To address inconsistent severity semantics without changing existing implementations, we introduce a lightweight severity measurement layer. For every degradation and each native severity level of a given backend, we quantify degradation strength using full reference image quality metrics: PSNR, SSIM, and LPIPS. This makes severity observable and comparable across sources while preserving native parameterizations. We demonstrate the framework through COCO Degradation, a taxonomy aligned benchmark for evaluating object detector robustness under diverse imaging conditions.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
