用相关性引导注意力，提升遥感变化检测精度与效率
CoSA: Correlation-Guided Change A ttention with Learnable Residual Gating for Remote Sensing Change Detection
- 设计相关性引导的变化注意力机制，结合可学习残差门控增强特征表达
- 在40张人工标注图像上达88.91% IoU，1000张伪标签样本平均置信度0.976
- 适用于少标注遥感图像变化检测，尤其适合监控摄像头场景迁移
像素级标注固定交通摄像头影像成本高昂，而从街景图像训练的模型在应用于高空监控时面临视角与外观显著偏移。本文提出一种数据高效的靶域流程：仅使用241张人工标注的CCTV图像和5,926张未标注帧。源域实验在3,300张第一人称视角（FPV）图像上训练31.0M参数自定义U-Net，获得93.05% IoU（FPV测试集330张）。该结果为基线，非迁移性能。释放的CCTV笔记本基于torchvision权重初始化42.0M参数DeepLabV3-ResNet50，未实现与U-Net检查点的兼容映射。在201张人工标注图像上训练并选取40张保留验证，获88.91% IoU。模型预测所有未标注帧，通过图像级置信度和最大连通区域先验排序，前1,000名样本平均置信度0.976，联合得分0.988。审计发现，报告的第二阶段98.52% IoU实测于150张仅含教师生成伪标签的子集；因目录布局错误，合并数据加载器未找到任何真实标签，将1,000个伪标签样本拆分为850训练+150评估。因此，98.52%应视为伪标签内部一致性而非真实标注准确率。可信的靶域结果为40张人工验证图像上的88.91% IoU。批量单次FP32推理在512×512分辨率下耗时12.98毫秒，对应77.03 FPS（NVIDIA RTX A6000 48GB GPU）。研究支持置信度与几何过滤的有效性，也揭示伪标签评估必须与自训练标签隔离。
原文摘要 · Abstract (English)
Pixel-level annotation of fixed traffic-camera imagery is expensive, while crosswalk models trained from street-level imagery face a substantial viewpoint and appearance shift when applied to elevated CCTV. We investigate a data-efficient target-domain pipeline using 241 manually annotated CCTV images and 5,926 unlabeled CCTV frames. A source-domain experiment trains a 31.0M-parameter custom U-Net on 3,300 first-person-view (FPV) images and obtains 93.05% IoU on its 330-image FPV test split. This result is a source baseline, not transferred performance: the released CCTV notebook instantiates a 42.0M-parameter DeepLabV3-ResNet50 from torchvision weights, and no compatible mapping from the U-Net checkpoint is implemented. Training on 201 manual CCTV images and selecting on 40 held-out manual masks yields 88.91% IoU. The model then predicts all unlabeled frames; image-level certainty and a largest-component area prior rank the candidates, and the top 1,000 attain mean certainty 0.976 and mean combined score 0.988. A repository audit shows that the reported second-stage 98.52% IoU was measured on a 150-image split containing only teacher-generated pseudo-masks. Because of a directory-layout mismatch, the executed combined-data loader found zero manual samples and split 1,000 pseudo-labeled samples into 850 training and 150 evaluation samples. We therefore report 98.52% as internal pseudo-label agreement rather than human-ground-truth accuracy. The defensible target-domain result is 88.91% IoU on the 40 manual validation images. Batch-one FP32 inference at 512 x 512 requires 12.98 ms, corresponding to 77.03 FPS, on an NVIDIA RTX A6000 48 GB GPU. These findings support the practicality of confidence-and-geometry filtering while also showing why pseudo-label evaluation must remain isolated from the labels used for self-training.
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。