{"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/how-to-extract-fashion-trends-from-social","title":"How To Extract Fashion Trends From Social Media? A Robust Object Detector With Support For Unsupervised Learning","arxiv_id":"1806.10787","date":"2018-06-28","proceeding":null,"authors":["Vijay Gabale","Anand Prabhu Subramanian"],"abstract":"With the proliferation of social media, fashion inspired from celebrities,\nreputed designers as well as fashion influencers has shortened the cycle of\nfashion design and manufacturing. However, with the explosion of fashion\nrelated content and large number of user generated fashion photos, it is an\narduous task for fashion designers to wade through social media photos and\ncreate a digest of trending fashion. This necessitates deep parsing of fashion\nphotos on social media to localize and classify multiple fashion items from a\ngiven fashion photo. While object detection competitions such as MSCOCO have\nthousands of samples for each of the object categories, it is quite difficult\nto get large labeled datasets for fast fashion items. Moreover,\nstate-of-the-art object detectors do not have any functionality to ingest large\namount of unlabeled data available on social media in order to fine tune object\ndetectors with labeled datasets. In this work, we show application of a generic\nobject detector, that can be pretrained in an unsupervised manner, on 24\ncategories from recently released Open Images V4 dataset. We first train the\nbase architecture of the object detector using unsupervisd learning on 60K\nunlabeled photos from 24 categories gathered from social media, and then\nsubsequently fine tune it on 8.2K labeled photos from Open Images V4 dataset.\nOn 300 X 300 image inputs, we achieve 72.7% mAP on a test dataset of 2.4K\nphotos while performing 11% to 17% better as compared to the state-of-the-art\nobject detectors. We show that this improvement is due to our choice of\narchitecture that lets us do unsupervised learning and that performs\nsignificantly better in identifying small objects.","url_abs":"http://arxiv.org/abs/1806.10787v1","url_pdf":"http://arxiv.org/pdf/1806.10787v1.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":"how-to-extract-fashion-trends-from-social","repo_url":"https://github.com/trhgu/awesome-fashion-contents","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"object","task_name":"Object"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/object-detection-on-sun-rgbd-val","task":"Object Detection","dataset":"SUN-RGBD val","model":"CDSSD","rank_in_archive_order":1,"of":1,"metrics":{"MAP":"7"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}