{"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/a-unified-model-with-structured-output-for","title":"A Unified Model with Structured Output for Fashion Images Classification","arxiv_id":"1806.09445","date":"2018-06-25","proceeding":null,"authors":["Beatriz Quintino Ferreira","Luís Baía","João Faria","Ricardo Gamelas Sousa"],"abstract":"A picture is worth a thousand words. Albeit a clich\\'e, for the fashion\nindustry, an image of a clothing piece allows one to perceive its category\n(e.g., dress), sub-category (e.g., day dress) and properties (e.g., white\ncolour with floral patterns). The seasonal nature of the fashion industry\ncreates a highly dynamic and creative domain with evermore data, making it\nunpractical to manually describe a large set of images (of products). In this\npaper, we explore the concept of visual recognition for fashion images through\nan end-to-end architecture embedding the hierarchical nature of the annotations\ndirectly into the model. Towards that goal, and inspired by the work of [7], we\nhave modified and adapted the original architecture proposal. Namely, we have\nremoved the message passing layer symmetry to cope with Farfetch category tree,\nadded extra layers for hierarchy level specificity, and moved the message\npassing layer into an enriched latent space. We compare the proposed unified\narchitecture against state-of-the-art models and demonstrate the performance\nadvantage of our model for structured multi-level categorization on a dataset\nof about 350k fashion product images.","url_abs":"http://arxiv.org/abs/1806.09445v1","url_pdf":"http://arxiv.org/pdf/1806.09445v1.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":"a-unified-model-with-structured-output-for","repo_url":"https://github.com/dpaddon/product_image_categorisation","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"specificity","task_name":"Specificity"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}