PulseAugur
实时 15:00:51
English(EN) Spatially Selective Self-Training for Unsupervised Building Change Detection

新框架改进无监督建筑变更检测

研究人员开发了一种名为SST-CD的新框架,用于使用遥感图像进行无监督建筑变更检测。该方法将问题重新表述为具有噪声伪监督的端到端检测器学习,重点关注由局部一致性标准识别的空间可靠像素。该框架还包含一个特征适配器和一个基于原型的解码器,以稳定训练并生成紧凑的表示。SST-CD在LEVIR-CD、WHU-CD和DSIFN-CD等基准数据集上展示了卓越的性能,优于现有的无标签方法。 AI

影响 增强了遥感分析的无监督学习能力,可能改进基础设施监测和城市规划。

排序理由 这是一篇详细介绍特定计算机视觉任务新方法的学术论文。

在 arXiv cs.CV 阅读 →

AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →

报道来源 [2]

  1. arXiv cs.CV TIER_1 English(EN) · Wafaa I. M. Hussin, Zhi Lu, Anas M. I. Mohammed, Xiang Zhou, Ratiba A. H. Abubaker, Zhenming Peng ·

    Spatially Selective Self-Training for Unsupervised Building Change Detection

    arXiv:2606.10775v1 Announce Type: new Abstract: Unsupervised building change detection aims to learn building-change masks from unlabeled bi-temporal remote sensing images. Existing label-free methods often follow a discrepancy-to-mask paradigm, directly using temporal difference…

  2. arXiv cs.CV TIER_1 English(EN) · Zhenming Peng ·

    Spatially Selective Self-Training for Unsupervised Building Change Detection

    Unsupervised building change detection aims to learn building-change masks from unlabeled bi-temporal remote sensing images. Existing label-free methods often follow a discrepancy-to-mask paradigm, directly using temporal differences, frozen foundation-model responses, prompt-bas…