{"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/siamese-network-based-metric-learning-for-sar","title":"SIAMESE NETWORK BASED METRIC LEARNING FOR SAR TARGET CLASSIFICATION","arxiv_id":null,"date":"2019-07-15","proceeding":"2019 IEEE 2019 7","authors":["Zongxu Pan1","2*","Xianjie Bao1","Yueting Zhang1","Bowei Wang1","Quanzhi An1","3","and Bin Lei1","2"],"abstract":"A Siamese network based metric learning method is \r\nproposed for SAR target classification with few training \r\nsamples. The network consists of two identical CNNs \r\nsharing the weights. Different from classification networks\r\nthat predict the category of one sample, the Siamese network\r\nimplements a metric learning to measure the similarity \r\nbetween two samples. Since the input is the sample pair, the \r\namount of training data dramatically increases which \r\ncontributes to training a better network. When generating the \r\npairs, a hard negative mining scheme is proposed for \r\nimproving the performance. To avoid computing the \r\nsimilarity between the test sample and each training sample \r\nat the test stage, which is time consuming, a two stages \r\nscheme is employed with an additional classification\r\nnetwork taking the output of the single branch of Siamese \r\nnetwork as the input and predicting the category.\r\nExperiments on the MSTAR dataset validate the \r\neffectiveness of the proposed method.","url_abs":"https://note.youdao.com/s/IztSKAI","url_pdf":"https://note.youdao.com/s/USopsPtp","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":"siamese-network-based-metric-learning-for-sar","repo_url":"https://github.com/tkxxfisb/SIAMESE-NETWORK-BASED-METRIC-LEARNING-FOR-SAR-TARGET-CLASSIFICATION","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"metric-learning","task_name":"Metric Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}