论文标题
一个基于变压器的暹罗网络用于变更检测
A Transformer-Based Siamese Network for Change Detection
论文作者
论文摘要
本文介绍了一对共同注册的遥感图像的基于变压器的暹罗网络体系结构（由变更形式的缩写），用于更改检测（CD）。与最近基于完全卷积网络（CORVNET）的CD框架不同，该提出的方法将层次结构化的变压器编码器与暹罗网络体系结构中的多层感知（MLP）解码器统一，以有效地渲染准确的CD所需的多尺度较长细节。两个CD数据集的实验表明，所提出的端到端可训练的变更形式体系结构的CD性能比以前的同行更好。我们的代码可在https://github.com/wgcban/changeformer上找到。
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.
