{"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/cliquecnn-deep-unsupervised-exemplar-learning","title":"CliqueCNN: Deep Unsupervised Exemplar Learning","arxiv_id":"1608.08792","date":"2016-08-31","proceeding":"NeurIPS 2016 12","authors":["Miguel A. Bautista","Artsiom Sanakoyeu","Ekaterina Sutter","Björn Ommer"],"abstract":"Exemplar learning is a powerful paradigm for discovering visual similarities\nin an unsupervised manner. In this context, however, the recent breakthrough in\ndeep learning could not yet unfold its full potential. With only a single\npositive sample, a great imbalance between one positive and many negatives, and\nunreliable relationships between most samples, training of Convolutional Neural\nnetworks is impaired. Given weak estimates of local distance we propose a\nsingle optimization problem to extract batches of samples with mutually\nconsistent relations. Conflicting relations are distributed over different\nbatches and similar samples are grouped into compact cliques. Learning exemplar\nsimilarities is framed as a sequence of clique categorization tasks. The CNN\nthen consolidates transitivity relations within and between cliques and learns\na single representation for all samples without the need for labels. The\nproposed unsupervised approach has shown competitive performance on detailed\nposture analysis and object classification.","url_abs":"http://arxiv.org/abs/1608.08792v1","url_pdf":"http://arxiv.org/pdf/1608.08792v1.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":"cliquecnn-deep-unsupervised-exemplar-learning","repo_url":"https://github.com/asanakoy/cliquecnn","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"BSD-2-Clause"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1608.08792","atlas_url":"https://app.syntology.ai/?focus=1608.08792","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}