{"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/finegan-unsupervised-hierarchical","title":"FineGAN: Unsupervised Hierarchical Disentanglement for Fine-Grained Object Generation and Discovery","arxiv_id":"1811.11155","date":"2018-11-27","proceeding":"CVPR 2019 6","authors":["Krishna Kumar Singh","Utkarsh Ojha","Yong Jae Lee"],"abstract":"We propose FineGAN, a novel unsupervised GAN framework, which disentangles\nthe background, object shape, and object appearance to hierarchically generate\nimages of fine-grained object categories. To disentangle the factors without\nsupervision, our key idea is to use information theory to associate each factor\nto a latent code, and to condition the relationships between the codes in a\nspecific way to induce the desired hierarchy. Through extensive experiments, we\nshow that FineGAN achieves the desired disentanglement to generate realistic\nand diverse images belonging to fine-grained classes of birds, dogs, and cars.\nUsing FineGAN's automatically learned features, we also cluster real images as\na first attempt at solving the novel problem of unsupervised fine-grained\nobject category discovery. Our code/models/demo can be found at\nhttps://github.com/kkanshul/finegan","url_abs":"http://arxiv.org/abs/1811.11155v2","url_pdf":"http://arxiv.org/pdf/1811.11155v2.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":"finegan-unsupervised-hierarchical","repo_url":"https://github.com/kkanshul/finegan","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"BSD-2-Clause"}}],"tasks":[{"task_slug":"conditional-image-generation","task_name":"Conditional Image Generation"},{"task_slug":"disentanglement","task_name":"Disentanglement"},{"task_slug":"fine-grained-visual-categorization","task_name":"Fine-Grained Visual Categorization"},{"task_slug":"image-clustering","task_name":"Image Clustering"},{"task_slug":"object","task_name":"Object"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-clustering-on-cub-birds","task":"Image Clustering","dataset":"CUB Birds","model":"FineGAN","rank_in_archive_order":1,"of":4,"metrics":{"Accuracy":"0.126","NMI":"0.403"},"uses_additional_data":false},{"leaderboard":"/sota/image-clustering-on-stanford-cars","task":"Image Clustering","dataset":"Stanford Cars","model":"FineGAN","rank_in_archive_order":2,"of":5,"metrics":{"Accuracy":"0.078","NMI":"0.354"},"uses_additional_data":false},{"leaderboard":"/sota/image-clustering-on-stanford-dogs","task":"Image Clustering","dataset":"Stanford Dogs","model":"FineGAN","rank_in_archive_order":1,"of":4,"metrics":{"Accuracy":"0.079","NMI":"0.233"},"uses_additional_data":false},{"leaderboard":"/sota/image-generation-on-cub-128-x-128","task":"Image Generation","dataset":"CUB 128 x 128","model":"FineGAN","rank_in_archive_order":2,"of":4,"metrics":{"FID":"11.25","Inception score":"52.53"},"uses_additional_data":false},{"leaderboard":"/sota/image-generation-on-stanford-cars","task":"Image Generation","dataset":"Stanford Cars","model":"FineGAN","rank_in_archive_order":2,"of":4,"metrics":{"FID":"16.03","Inception score":"32.62"},"uses_additional_data":false},{"leaderboard":"/sota/image-generation-on-stanford-dogs","task":"Image Generation","dataset":"Stanford Dogs","model":"FineGAN","rank_in_archive_order":2,"of":4,"metrics":{"FID":"25.66","Inception score":"46.92"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1811.11155","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}