{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/siamixformer-a-siamese-transformer-network","title":"SiamixFormer: a fully-transformer Siamese network with temporal Fusion for accurate building detection and change detection in bi-temporal remote sensing images","arxiv_id":"2208.00657","date":"2022-08-01","proceeding":null,"authors":["Amir mohammadian","Foad Ghaderi"],"abstract":"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.","url_abs":"https://arxiv.org/abs/2208.00657v2","url_pdf":"https://arxiv.org/pdf/2208.00657v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[],"tasks":[{"task_slug":"2d-semantic-segmentation","task_name":"2D Semantic Segmentation"},{"task_slug":"building-damage-assessment","task_name":"Building Damage Assessment"},{"task_slug":"building-change-detection-for-remote-sensing","task_name":"Building change detection for remote sensing images"},{"task_slug":"change-detection","task_name":"Change Detection"},{"task_slug":"change-detection-for-remote-sensing-images","task_name":"Change detection for remote sensing images"},{"task_slug":"decoder","task_name":"Decoder"},{"task_slug":"extracting-buildings-in-remote-sensing-images","task_name":"Extracting Buildings In Remote Sensing Images"}],"methods":[{"method_slug":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"mix-ffn","method_name":"Mix-FFN"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"segformer","method_name":"SegFormer"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"transformer","method_name":"Transformer"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/building-change-detection-for-remote-sensing","task":"Building change detection for remote sensing images","dataset":"LEVIR-CD","model":"SiamixFormer-5","rank_in_archive_order":13,"of":37,"metrics":{"F1":"91.58","IoU":"85.38"},"uses_additional_data":false},{"leaderboard":"/sota/building-change-detection-for-remote-sensing","task":"Building change detection for remote sensing images","dataset":"LEVIR-CD","model":"SiamixFormer-4","rank_in_archive_order":22,"of":37,"metrics":{"F1":"90.70","IoU":"84.05"},"uses_additional_data":false},{"leaderboard":"/sota/building-change-detection-for-remote-sensing","task":"Building change detection for remote sensing images","dataset":"LEVIR-CD","model":"SiamixFormer-2","rank_in_archive_order":23,"of":37,"metrics":{"F1":"90.57\t","IoU":"83.88"},"uses_additional_data":false},{"leaderboard":"/sota/building-change-detection-for-remote-sensing","task":"Building change detection for remote sensing images","dataset":"LEVIR-CD","model":"SiamixFormer-3","rank_in_archive_order":25,"of":37,"metrics":{"F1":"90.54","IoU":"83.84"},"uses_additional_data":false},{"leaderboard":"/sota/building-change-detection-for-remote-sensing","task":"Building change detection for remote sensing images","dataset":"LEVIR-CD","model":"SiamixFormer-0","rank_in_archive_order":31,"of":37,"metrics":{"F1":"89.47","IoU":"82.29"},"uses_additional_data":false},{"leaderboard":"/sota/building-change-detection-for-remote-sensing","task":"Building change detection for remote sensing images","dataset":"LEVIR-CD","model":"SiamixFormer-1","rank_in_archive_order":33,"of":37,"metrics":{"F1":"89.29","IoU":"82.03"},"uses_additional_data":false},{"leaderboard":"/sota/change-detection-for-remote-sensing-images-on","task":"Change detection for remote sensing images","dataset":"CDD Dataset (season-varying)","model":"SiamixFormer-5","rank_in_archive_order":8,"of":25,"metrics":{"F1-Score":"0.9713","IoU":"94.51"},"uses_additional_data":false},{"leaderboard":"/sota/change-detection-for-remote-sensing-images-on","task":"Change detection for remote sensing images","dataset":"CDD Dataset (season-varying)","model":"SiamixFormer-4","rank_in_archive_order":10,"of":25,"metrics":{"F1-Score":"0.9685","IoU":"94"},"uses_additional_data":false},{"leaderboard":"/sota/change-detection-for-remote-sensing-images-on","task":"Change detection for remote sensing images","dataset":"CDD Dataset (season-varying)","model":"SiamixFormer-3","rank_in_archive_order":12,"of":25,"metrics":{"F1-Score":"0.9648","IoU":"93.33"},"uses_additional_data":false},{"leaderboard":"/sota/change-detection-for-remote-sensing-images-on","task":"Change detection for remote sensing images","dataset":"CDD Dataset (season-varying)","model":"SiamixFormer-2","rank_in_archive_order":15,"of":25,"metrics":{"F1-Score":"0.9552","IoU":"91.62"},"uses_additional_data":false},{"leaderboard":"/sota/change-detection-for-remote-sensing-images-on","task":"Change detection for remote sensing images","dataset":"CDD Dataset (season-varying)","model":"SiamixFormer-1","rank_in_archive_order":20,"of":25,"metrics":{"F1-Score":"0.9278","IoU":"87"},"uses_additional_data":false},{"leaderboard":"/sota/change-detection-for-remote-sensing-images-on","task":"Change detection for remote sensing images","dataset":"CDD Dataset (season-varying)","model":"SiamixFormer-0","rank_in_archive_order":21,"of":25,"metrics":{"F1-Score":"0.9205","IoU":"85.81"},"uses_additional_data":false},{"leaderboard":"/sota/extracting-buildings-in-remote-sensing-images-2","task":"Extracting Buildings In Remote Sensing Images","dataset":"WHU Building Dataset","model":"SiamixFormer-5","rank_in_archive_order":1,"of":7,"metrics":{"F1":"96.69","IoU":"93.58"},"uses_additional_data":false},{"leaderboard":"/sota/extracting-buildings-in-remote-sensing-images-2","task":"Extracting Buildings In Remote Sensing Images","dataset":"WHU Building Dataset","model":"SiamixFormer-4","rank_in_archive_order":2,"of":7,"metrics":{"F1":"96.59","IoU":"93.4"},"uses_additional_data":false},{"leaderboard":"/sota/extracting-buildings-in-remote-sensing-images-2","task":"Extracting Buildings In Remote Sensing Images","dataset":"WHU Building Dataset","model":"SiamixFormer-3","rank_in_archive_order":3,"of":7,"metrics":{"F1":"96.32","IoU":"92.9"},"uses_additional_data":false},{"leaderboard":"/sota/extracting-buildings-in-remote-sensing-images-2","task":"Extracting Buildings In Remote Sensing Images","dataset":"WHU Building Dataset","model":"SiamixFormer-2","rank_in_archive_order":4,"of":7,"metrics":{"F1":"96.22","IoU":"92.7"},"uses_additional_data":false},{"leaderboard":"/sota/extracting-buildings-in-remote-sensing-images-2","task":"Extracting Buildings In Remote Sensing Images","dataset":"WHU Building Dataset","model":"SiamixFormer-1","rank_in_archive_order":5,"of":7,"metrics":{"F1":"96.01","IoU":"92.31"},"uses_additional_data":false},{"leaderboard":"/sota/extracting-buildings-in-remote-sensing-images-2","task":"Extracting Buildings In Remote Sensing Images","dataset":"WHU Building Dataset","model":"SiamixFormer-0","rank_in_archive_order":7,"of":7,"metrics":{"F1":"95.53","IoU":"91.43"},"uses_additional_data":false},{"leaderboard":"/sota/extracting-buildings-in-remote-sensing-images","task":"Extracting Buildings In Remote Sensing Images","dataset":"xBD","model":"SiamixFormer-5","rank_in_archive_order":1,"of":7,"metrics":{"F1":"88.43","IoU":"79.26"},"uses_additional_data":false},{"leaderboard":"/sota/extracting-buildings-in-remote-sensing-images","task":"Extracting Buildings In Remote Sensing Images","dataset":"xBD","model":"SiamixFormer-4","rank_in_archive_order":2,"of":7,"metrics":{"F1":"88.35","IoU":"79.14"},"uses_additional_data":false},{"leaderboard":"/sota/extracting-buildings-in-remote-sensing-images","task":"Extracting Buildings In Remote Sensing Images","dataset":"xBD","model":"SiamixFormer-3","rank_in_archive_order":3,"of":7,"metrics":{"F1":"88.3","IoU":"79.06"},"uses_additional_data":false},{"leaderboard":"/sota/extracting-buildings-in-remote-sensing-images","task":"Extracting Buildings In Remote Sensing Images","dataset":"xBD","model":"SiamixFormer-2","rank_in_archive_order":4,"of":7,"metrics":{"F1":"88.05","IoU":"78.66"},"uses_additional_data":false},{"leaderboard":"/sota/extracting-buildings-in-remote-sensing-images","task":"Extracting Buildings In Remote Sensing Images","dataset":"xBD","model":"SiamixFormer-1","rank_in_archive_order":5,"of":7,"metrics":{"F1":"87.23","IoU":"77.35"},"uses_additional_data":false},{"leaderboard":"/sota/extracting-buildings-in-remote-sensing-images","task":"Extracting Buildings In Remote Sensing Images","dataset":"xBD","model":"SiamixFormer-0","rank_in_archive_order":7,"of":7,"metrics":{"F1":"86.46","IoU":"76.43"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}