{"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/generate-to-adapt-aligning-domains-using","title":"Generate To Adapt: Aligning Domains using Generative Adversarial Networks","arxiv_id":"1704.01705","date":"2017-04-06","proceeding":"CVPR 2018 6","authors":["Swami Sankaranarayanan","Yogesh Balaji","Carlos D. Castillo","Rama Chellappa"],"abstract":"Domain Adaptation is an actively researched problem in Computer Vision. In\nthis work, we propose an approach that leverages unsupervised data to bring the\nsource and target distributions closer in a learned joint feature space. We\naccomplish this by inducing a symbiotic relationship between the learned\nembedding and a generative adversarial network. This is in contrast to methods\nwhich use the adversarial framework for realistic data generation and\nretraining deep models with such data. We demonstrate the strength and\ngenerality of our approach by performing experiments on three different tasks\nwith varying levels of difficulty: (1) Digit classification (MNIST, SVHN and\nUSPS datasets) (2) Object recognition using OFFICE dataset and (3) Domain\nadaptation from synthetic to real data. Our method achieves state-of-the art\nperformance in most experimental settings and by far the only GAN-based method\nthat has been shown to work well across different datasets such as OFFICE and\nDIGITS.","url_abs":"http://arxiv.org/abs/1704.01705v4","url_pdf":"http://arxiv.org/pdf/1704.01705v4.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":"generate-to-adapt-aligning-domains-using","repo_url":"https://github.com/watay147/tele_project","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"domain-adaptation","task_name":"Domain Adaptation"},{"task_slug":null,"task_name":"Generative Adversarial Network"},{"task_slug":"object-recognition","task_name":"Object Recognition"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/domain-adaptation-on-office-31","task":"Domain Adaptation","dataset":"Office-31","model":"GTA","rank_in_archive_order":27,"of":40,"metrics":{"Average Accuracy":"86.5"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1704.01705","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}