{"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/context-aware-embeddings-for-automatic-art","title":"Context-Aware Embeddings for Automatic Art Analysis","arxiv_id":"1904.04985","date":"2019-04-10","proceeding":null,"authors":["Noa Garcia","Benjamin Renoust","Yuta Nakashima"],"abstract":"Automatic art analysis aims to classify and retrieve artistic representations\nfrom a collection of images by using computer vision and machine learning\ntechniques. In this work, we propose to enhance visual representations from\nneural networks with contextual artistic information. Whereas visual\nrepresentations are able to capture information about the content and the style\nof an artwork, our proposed context-aware embeddings additionally encode\nrelationships between different artistic attributes, such as author, school, or\nhistorical period. We design two different approaches for using context in\nautomatic art analysis. In the first one, contextual data is obtained through a\nmulti-task learning model, in which several attributes are trained together to\nfind visual relationships between elements. In the second approach, context is\nobtained through an art-specific knowledge graph, which encodes relationships\nbetween artistic attributes. An exhaustive evaluation of both of our models in\nseveral art analysis problems, such as author identification, type\nclassification, or cross-modal retrieval, show that performance is improved by\nup to 7.3% in art classification and 37.24% in retrieval when context-aware\nembeddings are used.","url_abs":"http://arxiv.org/abs/1904.04985v1","url_pdf":"http://arxiv.org/pdf/1904.04985v1.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":"context-aware-embeddings-for-automatic-art","repo_url":"https://github.com/noagarcia/context-art-classification","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"art-analysis","task_name":"Art Analysis"},{"task_slug":"cross-modal-retrieval","task_name":"Cross-Modal Retrieval"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"multi-task-learning","task_name":"Multi-Task Learning"},{"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}