This paper presents a transformer-based Siamese network architecture (abbreviated by ChangeFormer) for Change Detection (CD) from a pair of co-registered remote sensing images. Different from recent CD frameworks, which are based on fully convolutional networks (ConvNets), the proposed method unifies hierarchically structured transformer encoder with Multi-Layer Perception (MLP) decoder in a Siamese network architecture to efficiently render multi-scale long-range details required for accurate CD. Experiments on two CD datasets show that the proposed end-to-end trainable ChangeFormer architecture achieves better CD performance than previous counterparts. Our code is available at https://github.com/wgcban/ChangeFormer.

翻译：本文件介绍了基于变压器的Siamse网络结构(由CreateFormer提供),用于变化探测(CD),由一对共同注册的遥感图像组成。与最近的CD框架不同,后者基于全演化网络(ConvNets),拟议方法将结构分层的变压器编码器与一个Siamesie网络结构的多重感知解码器(MLP)统一起来,以便有效地提供准确的CD所需的多尺度长距离细节。对两个CD数据集的实验显示,拟议的端到端可训练的变换Former结构的CD性能优于以前的对等结构。我们的代码可在https://github.com/wgcban/ChangeFormer查阅。