# SiamixFormer: a fully-transformer Siamese network with temporal Fusion for accurate building detection and change detection in bi-temporal remote sensing images

arXiv: https://arxiv.org/abs/2208.00657 · HTML: https://mlsanity.com/paper/2208.00657/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- Builds on 16 of its 46 references that are in the map.
- First posted on arXiv 1 Aug 2022.

## Builds on (most cited first, 16 of 16)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Gaussian Error Linear Units (GELUs)](https://mlsanity.com/paper/1606.08415/) (2016)
- [SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers](https://mlsanity.com/paper/2105.15203/) (2021)
- [CBAM: Convolutional Block Attention Module](https://mlsanity.com/paper/1807.06521/) (2018)
- [Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions](https://mlsanity.com/paper/2102.12122/) (2021)
- [Focal Loss for Dense Object Detection](https://mlsanity.com/paper/1708.02002/) (2017)
- [Generalised Dice overlap as a deep learning loss function for highly unbalanced segmentations](https://mlsanity.com/paper/1707.03237/) (2017)
- [Remote Sensing Image Change Detection with Transformers](https://mlsanity.com/paper/2103.00208/) (2021)
- [Tackling Climate Change with Machine Learning](https://mlsanity.com/paper/1906.05433/) (2019)
- [A Transformer-Based Siamese Network for Change Detection](https://mlsanity.com/paper/2201.01293/) (2022)
- [An Empirical Study of Remote Sensing Pretraining](https://mlsanity.com/paper/2204.02825/) (2022)
- [Building Change Detection for Remote Sensing Images Using a Dual Task Constrained Deep Siamese Convolutional Network Model](https://mlsanity.com/paper/1909.07726/) (2019)
- [Dual-Tasks Siamese Transformer Framework for Building Damage Assessment](https://mlsanity.com/paper/2201.10953/) (2022)
- [BDANet: Multiscale Convolutional Neural Network with Cross-directional Attention for Building Damage Assessment from Satellite Images](https://mlsanity.com/paper/2105.07364/) (2021)
- [SiamTrans: Zero-Shot Multi-Frame Image Restoration with Pre-Trained Siamese Transformers](https://mlsanity.com/paper/2112.09426/) (2021)

## Built on by (most cited first, 10 of 10)

- [MTP: Advancing Remote Sensing Foundation Model via Multi-Task Pretraining](https://mlsanity.com/paper/2403.13430/) (2024)
- [VcT: Visual change Transformer for Remote Sensing Image Change Detection](https://mlsanity.com/paper/2310.11417/) (2023)
- [Visual Instruction Pretraining for Domain-Specific Foundation Models](https://mlsanity.com/paper/2509.17562/) (2025)
- [DamageCAT: A Deep Learning Transformer Framework for Typology-Based Post-Disaster Building Damage Categorization](https://mlsanity.com/paper/2504.11637/) (2025)
- [SRC-Net: Bi-Temporal Spatial Relationship Concerned Network for Change Detection](https://mlsanity.com/paper/2406.05668/) (2024)
- [Recov-Vision: Linking Street View Imagery and Vision-Language Models for Post-Disaster Recovery](https://mlsanity.com/paper/2509.20628/) (2025)
- [Optimizing Latent Representations for Robust Building Damage Assessment Onboard Earth Observation Satellites](https://mlsanity.com/paper/2605.29575/) (2026)
- [Damage-TriageFormer: A Foundation-Model Framework for Typology-Based Building Damage Assessment from Mono-Temporal Imagery](https://mlsanity.com/paper/2606.12248/) (2026)
- [Embedded Bi-Temporal Building Damage Assessment for On-Board Data Reduction](https://mlsanity.com/paper/2609.37013/) (2026)
- [MSSFC-Net:Enhancing Building Interpretation with Multi-Scale Spatial-Spectral Feature Collaboration](https://mlsanity.com/paper/2504.00759/) (2025)

## Abstract (arXiv)

Building detection and change detection using remote sensing images can help urban and rescue planning. Moreover, they can be used for building damage assessment after natural disasters. Currently, most of the existing models for building detection use only one image (pre-disaster image) to detect buildings. This is based on the idea that post-disaster images reduce the model's performance because of presence of destroyed buildings. In this paper, we propose a siamese model, called SiamixFormer, which uses pre- and post-disaster images as input. Our model has two encoders and has a hierarchical transformer architecture. The output of each stage in both encoders is given to a temporal transformer for feature fusion in a way that query is generated from pre-disaster images and (key, value) is generated from post-disaster images. To this end, temporal features are also considered in feature fusion. Another advantage of using temporal transformers in feature fusion is that they can better maintain large receptive fields generated by transformer encoders compared with CNNs. Finally, the output of the temporal transformer is given to a simple MLP decoder at each stage. The SiamixFormer model is evaluated on xBD, and WHU datasets, for building detection and on LEVIR-CD and CDD datasets for change detection and could outperform the state-of-the-art.