{"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/metric-learning-for-generalizing-spatial","title":"Metric Learning for Generalizing Spatial Relations to New Objects","arxiv_id":"1703.01946","date":"2017-03-06","proceeding":null,"authors":["Oier Mees","Nichola Abdo","Mladen Mazuran","Wolfram Burgard"],"abstract":"Human-centered environments are rich with a wide variety of spatial relations\nbetween everyday objects. For autonomous robots to operate effectively in such\nenvironments, they should be able to reason about these relations and\ngeneralize them to objects with different shapes and sizes. For example, having\nlearned to place a toy inside a basket, a robot should be able to generalize\nthis concept using a spoon and a cup. This requires a robot to have the\nflexibility to learn arbitrary relations in a lifelong manner, making it\nchallenging for an expert to pre-program it with sufficient knowledge to do so\nbeforehand. In this paper, we address the problem of learning spatial relations\nby introducing a novel method from the perspective of distance metric learning.\nOur approach enables a robot to reason about the similarity between pairwise\nspatial relations, thereby enabling it to use its previous knowledge when\npresented with a new relation to imitate. We show how this makes it possible to\nlearn arbitrary spatial relations from non-expert users using a small number of\nexamples and in an interactive manner. Our extensive evaluation with real-world\ndata demonstrates the effectiveness of our method in reasoning about a\ncontinuous spectrum of spatial relations and generalizing them to new objects.","url_abs":"http://arxiv.org/abs/1703.01946v3","url_pdf":"http://arxiv.org/pdf/1703.01946v3.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":"metric-learning-for-generalizing-spatial","repo_url":"https://github.com/mees/generalize_spatial_relations","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"metric-learning","task_name":"Metric Learning"}],"methods":[],"datasets_introduced":[{"slug":"freiburg-spatial-relations","name":"Freiburg Spatial Relations","full_name":"Freiburg Spatial Relations"}],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}