{"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/all-graphs-lead-to-rome-learning-geometric","title":"All Graphs Lead to Rome: Learning Geometric and Cycle-Consistent Representations with Graph Convolutional Networks","arxiv_id":"1901.02078","date":"2019-01-07","proceeding":null,"authors":["Stephen Phillips","Kostas Daniilidis"],"abstract":"Image feature matching is a fundamental part of many geometric computer\nvision applications, and using multiple images can improve performance. In this\nwork, we formulate multi-image matching as a graph embedding problem then use a\nGraph Convolutional Network to learn an appropriate embedding function for\naligning image features. We use cycle consistency to train our network in an\nunsupervised fashion, since ground truth correspondence is difficult or\nexpensive to aquire. In addition, geometric consistency losses can be added at\ntraining time, even if the information is not available in the test set, unlike\nprevious approaches that optimize cycle consistency directly. To the best of\nour knowledge, no other works have used learning for multi-image feature\nmatching. Our experiments show that our method is competitive with other\noptimization based approaches.","url_abs":"http://arxiv.org/abs/1901.02078v1","url_pdf":"http://arxiv.org/pdf/1901.02078v1.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":"all-graphs-lead-to-rome-learning-geometric","repo_url":"https://github.com/daniilidis-group/all-graphs-lead-to-rome","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"all","task_name":"All"},{"task_slug":"graph-embedding","task_name":"Graph Embedding"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}