{"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/object-category-learning-and-retrieval-with","title":"Object category learning and retrieval with weak supervision","arxiv_id":"1801.08985","date":"2018-01-26","proceeding":null,"authors":["Steven Hickson","Anelia Angelova","Irfan Essa","Rahul Sukthankar"],"abstract":"We consider the problem of retrieving objects from image data and learning to\nclassify them into meaningful semantic categories with minimal supervision. To\nthat end, we propose a fully differentiable unsupervised deep clustering\napproach to learn semantic classes in an end-to-end fashion without individual\nclass labeling using only unlabeled object proposals. The key contributions of\nour work are 1) a kmeans clustering objective where the clusters are learned as\nparameters of the network and are represented as memory units, and 2)\nsimultaneously building a feature representation, or embedding, while learning\nto cluster it. This approach shows promising results on two popular computer\nvision datasets: on CIFAR10 for clustering objects, and on the more complex and\nchallenging Cityscapes dataset for semantically discovering classes which\nvisually correspond to cars, people, and bicycles. Currently, the only\nsupervision provided is segmentation objectness masks, but this method can be\nextended to use an unsupervised objectness-based object generation mechanism\nwhich will make the approach completely unsupervised.","url_abs":"http://arxiv.org/abs/1801.08985v2","url_pdf":"http://arxiv.org/pdf/1801.08985v2.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":"object-category-learning-and-retrieval-with","repo_url":"https://github.com/StevenHickson/VideoObjectProposals","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":null}],"tasks":[{"task_slug":"clustering","task_name":"Clustering"},{"task_slug":"deep-clustering","task_name":"Deep Clustering"},{"task_slug":"object","task_name":"Object"},{"task_slug":"retrieval","task_name":"Retrieval"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}