{"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/supervised-learning-of-semantics-preserving","title":"Supervised Learning of Semantics-Preserving Hash via Deep Convolutional Neural Networks","arxiv_id":"1507.00101","date":"2015-07-01","proceeding":null,"authors":["Huei-Fang Yang","Kevin Lin","Chu-Song Chen"],"abstract":"This paper presents a simple yet effective supervised deep hash approach that\nconstructs binary hash codes from labeled data for large-scale image search. We\nassume that the semantic labels are governed by several latent attributes with\neach attribute on or off, and classification relies on these attributes. Based\non this assumption, our approach, dubbed supervised semantics-preserving deep\nhashing (SSDH), constructs hash functions as a latent layer in a deep network\nand the binary codes are learned by minimizing an objective function defined\nover classification error and other desirable hash codes properties. With this\ndesign, SSDH has a nice characteristic that classification and retrieval are\nunified in a single learning model. Moreover, SSDH performs joint learning of\nimage representations, hash codes, and classification in a point-wised manner,\nand thus is scalable to large-scale datasets. SSDH is simple and can be\nrealized by a slight enhancement of an existing deep architecture for\nclassification; yet it is effective and outperforms other hashing approaches on\nseveral benchmarks and large datasets. Compared with state-of-the-art\napproaches, SSDH achieves higher retrieval accuracy, while the classification\nperformance is not sacrificed.","url_abs":"http://arxiv.org/abs/1507.00101v2","url_pdf":"http://arxiv.org/pdf/1507.00101v2.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":"supervised-learning-of-semantics-preserving","repo_url":"https://github.com/kevinlin311tw/Caffe-DeepBinaryCode","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"attribute","task_name":"Attribute"},{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"deep-hashing","task_name":"Deep Hashing"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"image-retrieval","task_name":"Image Retrieval"},{"task_slug":"retrieval","task_name":"Retrieval"}],"methods":[{"method_slug":"affine-coupling","method_name":"Affine Coupling"},{"method_slug":"normalizing-flows","method_name":"Normalizing Flows"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}