无需训练即可检测任意类别变化，提升遥感图像变化检测泛化能力。
AdaptOVCD: Training-Free Open-Vocabulary Remote Sensing Change Detection via Adaptive Information Fusion
- 通过多层级自适应融合，整合辐射、特征与决策信息提升检测精度。
- 零样本下跨数据集性能达全监督上限的84.89%，显著优于现有方法。
- 适合需要快速部署、无标注数据的遥感变化检测场景。
遥感变化检测在环境监测、城市规划和灾害评估中至关重要。然而，现有方法通常依赖预定义类别和大规模像素级标注，限制了其在开放世界中的泛化能力。为此，本文提出 AdaptOVCD，一种基于双维多层级信息融合的训练免费开放词汇变化检测架构。该框架在数据、特征和决策三个层面实现垂直融合，并引入水平自适应设计，增强异构预训练模型间的协同，有效抑制误差传播。具体而言：(1) 数据层采用自适应辐射校准（ARA），融合辐射统计与原始纹理特征，并与 SAM-HQ 配合实现辐射一致性分割；(2) 特征层采用自适应变化阈值（ACT），结合全局差异分布与边缘结构先验，利用 DINOv3 实现鲁棒变化检测；(3) 决策层采用自适应置信度过滤（ACF），融合语义置信度与空间约束，协同 DGTRS-CLIP 实现高置信度语义识别。在九种场景的综合评估中，AdaptOVCD 实现任意类别变化的零样本检测，显著超越现有训练免费方法。跨数据集评估中，其性能达到全监督上限的 84.89%，展现出优异泛化能力。代码已开源：https://github.com/Dmygithub/AdaptOVCD。
原文摘要 · Abstract (English)
Remote sensing change detection plays a pivotal role in domains such as environmental monitoring, urban planning, and disaster assessment. However, existing methods typically rely on predefined categories and large-scale pixel-level annotations, which limit their generalization and applicability in open-world scenarios. To address these limitations, this paper proposes AdaptOVCD, a training-free Open-Vocabulary Change Detection (OVCD) architecture based on dual-dimensional multi-level information fusion. The framework integrates multi-level information fusion across data, feature, and decision levels vertically while incorporating targeted adaptive designs horizontally, achieving deep synergy among heterogeneous pre-trained models to effectively mitigate error propagation. Specifically, (1) at the data level, Adaptive Radiometric Alignment (ARA) fuses radiometric statistics with original texture features and synergizes with SAM-HQ to achieve radiometrically consistent segmentation; (2) at the feature level, Adaptive Change Thresholding (ACT) combines global difference distributions with edge structure priors and leverages DINOv3 to achieve robust change detection; (3) at the decision level, Adaptive Confidence Filtering (ACF) integrates semantic confidence with spatial constraints and collaborates with DGTRS-CLIP to achieve high-confidence semantic identification. Comprehensive evaluations across nine scenarios demonstrate that AdaptOVCD detects arbitrary category changes in a zero-shot manner, significantly outperforming existing training-free methods. Meanwhile, it achieves 84.89\% of the fully-supervised performance upper bound in cross-dataset evaluations and exhibits superior generalization capabilities. The code is available at https://github.com/Dmygithub/AdaptOVCD.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。