{"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/mining-on-manifolds-metric-learning-without","title":"Mining on Manifolds: Metric Learning without Labels","arxiv_id":"1803.11095","date":"2018-03-29","proceeding":"CVPR 2018 6","authors":["Ahmet Iscen","Giorgos Tolias","Yannis Avrithis","Ondrej Chum"],"abstract":"In this work we present a novel unsupervised framework for hard training\nexample mining. The only input to the method is a collection of images relevant\nto the target application and a meaningful initial representation, provided\ne.g. by pre-trained CNN. Positive examples are distant points on a single\nmanifold, while negative examples are nearby points on different manifolds.\nBoth types of examples are revealed by disagreements between Euclidean and\nmanifold similarities. The discovered examples can be used in training with any\ndiscriminative loss. The method is applied to unsupervised fine-tuning of\npre-trained networks for fine-grained classification and particular object\nretrieval. Our models are on par or are outperforming prior models that are\nfully or partially supervised.","url_abs":"http://arxiv.org/abs/1803.11095v1","url_pdf":"http://arxiv.org/pdf/1803.11095v1.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":"mining-on-manifolds-metric-learning-without","repo_url":"https://github.com/gtolias/mom","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"metric-learning","task_name":"Metric Learning"},{"task_slug":"retrieval","task_name":"Retrieval"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1803.11095","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}