HSANet通过混合注意力机制提升遥感变化检测精度。
HSANET: A Hybrid Self-Cross Attention Network For Remote Sensing Change Detection
- 分层卷积提取多尺度特征，融合自注意力与跨尺度注意力
- 在两个公开数据集上达到最优性能，边缘细节更清晰
- 适合遥感图像变化检测任务，代码已开源
遥感图像变化检测是大规模监测的重要手段。本文提出HSANet，采用分层卷积提取多尺度特征，并引入混合自注意力与跨注意力机制，以学习和融合全局及跨尺度信息。该设计使模型能在不同尺度下捕捉全局上下文，并整合跨尺度特征，从而优化边缘细节，提升检测性能。相关代码已开源：https://github.com/ChengxiHAN/HSANet。
原文摘要 · Abstract (English)
The remote sensing image change detection task is an essential method for large-scale monitoring. We propose HSANet, a network that uses hierarchical convolution to extract multi-scale features. It incorporates hybrid self-attention and cross-attention mechanisms to learn and fuse global and cross-scale information. This enables HSANet to capture global context at different scales and integrate cross-scale features, refining edge details and improving detection performance. We will also open-source our model code: https://github.com/ChengxiHAN/HSANet.
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。