{"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/deep-micro-dictionary-learning-and-coding","title":"Deep Micro-Dictionary Learning and Coding Network","arxiv_id":"1809.04185","date":"2018-09-11","proceeding":null,"authors":["Hao Tang","Heng Wei","Wei Xiao","Wei Wang","Dan Xu","Yan Yan","Nicu Sebe"],"abstract":"In this paper, we propose a novel Deep Micro-Dictionary Learning and Coding\nNetwork (DDLCN). DDLCN has most of the standard deep learning layers (pooling,\nfully, connected, input/output, etc.) but the main difference is that the\nfundamental convolutional layers are replaced by novel compound dictionary\nlearning and coding layers. The dictionary learning layer learns an\nover-complete dictionary for the input training data. At the deep coding layer,\na locality constraint is added to guarantee that the activated dictionary bases\nare close to each other. Next, the activated dictionary atoms are assembled\ntogether and passed to the next compound dictionary learning and coding layers.\nIn this way, the activated atoms in the first layer can be represented by the\ndeeper atoms in the second dictionary. Intuitively, the second dictionary is\ndesigned to learn the fine-grained components which are shared among the input\ndictionary atoms. In this way, a more informative and discriminative low-level\nrepresentation of the dictionary atoms can be obtained. We empirically compare\nthe proposed DDLCN with several dictionary learning methods and deep learning\narchitectures. The experimental results on four popular benchmark datasets\ndemonstrate that the proposed DDLCN achieves competitive results compared with\nstate-of-the-art approaches.","url_abs":"http://arxiv.org/abs/1809.04185v2","url_pdf":"http://arxiv.org/pdf/1809.04185v2.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":"deep-micro-dictionary-learning-and-coding","repo_url":"https://github.com/Ha0Tang/DDLCN","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"dictionary-learning","task_name":"Dictionary Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}