{"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/robust-angular-local-descriptor-learning","title":"Robust Angular Local Descriptor Learning","arxiv_id":"1901.07076","date":"2019-01-21","proceeding":null,"authors":["Yanwu Xu","Mingming Gong","Tongliang Liu","Kayhan Batmanghelich","Chaohui Wang"],"abstract":"In recent years, the learned local descriptors have outperformed handcrafted\nones by a large margin, due to the powerful deep convolutional neural network\narchitectures such as L2-Net [1] and triplet based metric learning [2].\nHowever, there are two problems in the current methods, which hinders the\noverall performance. Firstly, the widely-used margin loss is sensitive to\nincorrect correspondences, which are prevalent in the existing local descriptor\nlearning datasets. Second, the L2 distance ignores the fact that the feature\nvectors have been normalized to unit norm. To tackle these two problems and\nfurther boost the performance, we propose a robust angular loss which 1) uses\ncosine similarity instead of L2 distance to compare descriptors and 2) relies\non a robust loss function that gives smaller penalty to triplets with negative\nrelative similarity. The resulting descriptor shows robustness on different\ndatasets, reaching the state-of-the-art result on Brown dataset , as well as\ndemonstrating excellent generalization ability on the Hpatches dataset and a\nWide Baseline Stereo dataset.","url_abs":"http://arxiv.org/abs/1901.07076v2","url_pdf":"http://arxiv.org/pdf/1901.07076v2.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":"robust-angular-local-descriptor-learning","repo_url":"https://github.com/xuyanwu/RAL-Net","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"metric-learning","task_name":"Metric Learning"},{"task_slug":null,"task_name":"Triplet"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1901.07076","atlas_url":"https://app.syntology.ai/?focus=1901.07076","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}