发表机构
Dalian Maritime University; School of Electronic Information and Artificial Intelligence, Shaanxi University of Science and Technology; School of Robotics, Hunan University; Department of Electronic and Electrical Engineering, Brunel University of London（大连海事大学; 陕西科技大学电子信息与人工智能学院; 湖南大学机器人学院; 布鲁内尔伦敦大学电子与电气工程系）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对点监督变化检测的伪标签不完整且含噪问题，提出引入SAM2先验的两阶段渐进式优化框架，在三类基准数据集上表现优于多数弱监督方法且具全监督方法竞争力。
AI 中文摘要
点监督变化检测（PS-CD）旨在仅利用稀疏标注点识别双时相图像间的像素级变化。尽管点标注大幅降低了标注成本，但其有限的空间覆盖往往导致伪标签不完整且存在噪声。为解决该问题，本文提出一种两阶段框架，将SAM2先验引入PS-CD并逐步使其适配目标任务。第一阶段，SAM2从双时相图像的点标注生成感知对象的候选掩码，设计双时相掩码选择策略将通用分割响应转换为更可靠的变化伪标签；随后采用带不确定性感知损失的轻量CNN细化模块，提升边界质量与局部结构一致性。第二阶段，构建师生自训练框架，其中教师模型通过指数移动平均更新并定期刷新伪标签，该设计建立了在伪标签细化与模型再优化间交替的闭环优化过程。在WHU-CD、LEVIR-CD、SYSU-CD三个基准数据集上的实验表明，所提方法在多数基准上优于现有弱监督方法，且与若干全监督方法相比仍具竞争力。
英文摘要
Point-supervised change detection (PS-CD) aims to identify pixel-level changes between bi-temporal images using only sparsely annotated points. Although point annotations substantially reduce labeling costs, their limited spatial coverage often results in incomplete and noisy pseudo-labels. To address this issue, we propose a two-stage framework that introduces SAM2 priors into PS-CD and progressively adapts them to the target task. In Stage I, SAM2 generates object-aware candidate masks from point annotations on the bi-temporal images, and a bi-temporal mask selection strategy is designed to convert generic segmentation responses into more reliable change pseudo-labels. Subsequently, a lightweight CNN refinement module with an uncertainty-aware loss is employed to improve boundary quality and local structural consistency. In Stage II, we construct a teacher-student self-training framework in which the teacher is updated by exponential moving average and periodically refreshes the pseudo-labels. This design establishes a closed-loop optimization process that alternates between pseudo-label refinement and model re-optimization. Experiments on three benchmark datasets, including WHU-CD, LEVIR-CD, and SYSU-CD, demonstrate that the proposed method outperforms previous weakly supervised approaches on most benchmarks and remains competitive with several fully supervised methods.
Comments9 pages,4 figures
