{"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/fast-low-rank-shared-dictionary-learning-for","title":"Fast Low-rank Shared Dictionary Learning for Image Classification","arxiv_id":"1610.08606","date":"2016-10-27","proceeding":null,"authors":["Tiep Vu","Vishal Monga"],"abstract":"Despite the fact that different objects possess distinct class-specific\nfeatures, they also usually share common patterns. This observation has been\nexploited partially in a recently proposed dictionary learning framework by\nseparating the particularity and the commonality (COPAR). Inspired by this, we\npropose a novel method to explicitly and simultaneously learn a set of common\npatterns as well as class-specific features for classification with more\nintuitive constraints. Our dictionary learning framework is hence characterized\nby both a shared dictionary and particular (class-specific) dictionaries. For\nthe shared dictionary, we enforce a low-rank constraint, i.e. claim that its\nspanning subspace should have low dimension and the coefficients corresponding\nto this dictionary should be similar. For the particular dictionaries, we\nimpose on them the well-known constraints stated in the Fisher discrimination\ndictionary learning (FDDL). Further, we develop new fast and accurate\nalgorithms to solve the subproblems in the learning step, accelerating its\nconvergence. The said algorithms could also be applied to FDDL and its\nextensions. The efficiencies of these algorithms are theoretically and\nexperimentally verified by comparing their complexities and running time with\nthose of other well-known dictionary learning methods. Experimental results on\nwidely used image datasets establish the advantages of our method over\nstate-of-the-art dictionary learning methods.","url_abs":"http://arxiv.org/abs/1610.08606v3","url_pdf":"http://arxiv.org/pdf/1610.08606v3.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":"fast-low-rank-shared-dictionary-learning-for","repo_url":"https://github.com/tiepvupsu/DICTOL","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"fast-low-rank-shared-dictionary-learning-for","repo_url":"https://github.com/tiepvupsu/DICTOL_python","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"dictionary-learning","task_name":"Dictionary Learning"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"image-classification","task_name":"image-classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1610.08606","atlas_url":"https://app.syntology.ai/?focus=1610.08606","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}