{"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/farseg-foreground-aware-relation-network-for","title":"FarSeg++: Foreground-Aware Relation Network for Geospatial Object Segmentation in High Spatial Resolution Remote Sensing Imagery","arxiv_id":null,"date":"2023-07-13","proceeding":"IEEE Transactions on Pattern Analysis and Machine Intelligence 2023 7","authors":["Zhuo Zheng","Yanfei Zhong","Junjue Wang","Ailong Ma","Liangpei Zhang"],"abstract":"Geospatial object segmentation, a fundamental Earth vision task, always suffers from scale variation, the larger intraclass variance of background, and foreground-background imbalance in high spatial resolution (HSR) remote sensing imagery. Generic semantic segmentation methods mainly focus on the scale variation in natural scenarios. However, the other two problems are insufficiently considered in large area Earth observation scenarios. In this paper, we propose a foreground-aware relation network (FarSeg++) from the perspectives of relation-based, optimizationbased, and objectness-based foreground modeling, alleviating the above two problems. From the perspective of the relations, the foreground-scene relation module improves the discrimination of the foreground features via the foreground-correlated contexts associated with the object-scene relation. From the perspective of\r\noptimization, foreground-aware optimization is proposed to focus on foreground examples and hard examples of the background during training to achieve a balanced optimization. Besides, from the perspective of objectness, a foreground-aware decoder is proposed to improve the objectness representation, alleviating the objectness prediction problem that is the main bottleneck revealed by an empirical upper bound analysis. We also introduce a new large-scale high-resolution urban vehicle segmentation dataset to verify the effectiveness of the proposed method and push the development of objectness prediction further forward. The experimental results suggest that FarSeg++ is superior to the state-of-the-art generic semantic segmentation methods and can achieve a better trade-off between speed and accuracy.","url_abs":"https://ieeexplore.ieee.org/document/10188509","url_pdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10188509","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":[{"paper_slug":"farseg-foreground-aware-relation-network-for","repo_url":"https://github.com/Z-Zheng/FarSeg","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"earth-observation","task_name":"Earth Observation"},{"task_slug":null,"task_name":"Relation"},{"task_slug":"relation-network","task_name":"Relation Network"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"the-semantic-segmentation-of-remote-sensing","task_name":"The Semantic Segmentation Of Remote Sensing Imagery"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"focus","method_name":"Focus"},{"method_slug":"speed","method_name":"SPEED"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/semantic-segmentation-on-isaid","task":"Semantic Segmentation","dataset":"iSAID","model":"FarSeg++@MiT-B2","rank_in_archive_order":7,"of":19,"metrics":{"mIoU":"67.9"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-isaid","task":"Semantic Segmentation","dataset":"iSAID","model":"FarSeg++@ResNet-50","rank_in_archive_order":8,"of":19,"metrics":{"mIoU":"67.6"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-isaid","task":"Semantic Segmentation","dataset":"iSAID","model":"FarSeg++@Swin-T","rank_in_archive_order":11,"of":19,"metrics":{"mIoU":"66.3"},"uses_additional_data":false},{"leaderboard":"/sota/the-semantic-segmentation-of-remote-sensing","task":"The Semantic Segmentation Of Remote Sensing Imagery","dataset":"UV6K","model":"FarSeg++@MiT-B2","rank_in_archive_order":1,"of":3,"metrics":{"IoU (%)":"66.4"},"uses_additional_data":false},{"leaderboard":"/sota/the-semantic-segmentation-of-remote-sensing","task":"The Semantic Segmentation Of Remote Sensing Imagery","dataset":"UV6K","model":"FarSeg++@Swin-T","rank_in_archive_order":2,"of":3,"metrics":{"IoU (%)":"65.8"},"uses_additional_data":false},{"leaderboard":"/sota/the-semantic-segmentation-of-remote-sensing","task":"The Semantic Segmentation Of Remote Sensing Imagery","dataset":"UV6K","model":"FarSeg++@ResNet-50","rank_in_archive_order":3,"of":3,"metrics":{"IoU (%)":"64.4"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}