{"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/semantic-labeling-in-very-high-resolution","title":"Semantic Labeling in Very High Resolution Images via a Self-Cascaded Convolutional Neural Network","arxiv_id":"1807.11236","date":"2018-07-30","proceeding":null,"authors":["Yongcheng Liu","Bin Fan","Lingfeng Wang","Jun Bai","Shiming Xiang","Chunhong Pan"],"abstract":"Semantic labeling for very high resolution (VHR) images in urban areas, is of\nsignificant importance in a wide range of remote sensing applications. However,\nmany confusing manmade objects and intricate fine-structured objects make it\nvery difficult to obtain both coherent and accurate labeling results. For this\nchallenging task, we propose a novel deep model with convolutional neural\nnetworks (CNNs), i.e., an end-to-end self-cascaded network (ScasNet).\nSpecifically, for confusing manmade objects, ScasNet improves the labeling\ncoherence with sequential global-to-local contexts aggregation. Technically,\nmulti-scale contexts are captured on the output of a CNN encoder, and then they\nare successively aggregated in a self-cascaded manner. Meanwhile, for\nfine-structured objects, ScasNet boosts the labeling accuracy with a\ncoarse-to-fine refinement strategy. It progressively refines the target objects\nusing the low-level features learned by CNN's shallow layers. In addition, to\ncorrect the latent fitting residual caused by multi-feature fusion inside\nScasNet, a dedicated residual correction scheme is proposed. It greatly\nimproves the effectiveness of ScasNet. Extensive experimental results on three\npublic datasets, including two challenging benchmarks, show that ScasNet\nachieves the state-of-the-art performance.","url_abs":"http://arxiv.org/abs/1807.11236v1","url_pdf":"http://arxiv.org/pdf/1807.11236v1.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":[{"paper_slug":"semantic-labeling-in-very-high-resolution","repo_url":"https://github.com/Yochengliu/ScasNet","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}