{"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/unsupervised-holistic-image-generation-from","title":"Unsupervised Holistic Image Generation from Key Local Patches","arxiv_id":"1703.10730","date":"2017-03-31","proceeding":"ECCV 2018 9","authors":["Donghoon Lee","Sangdoo Yun","Sungjoon Choi","Hwiyeon Yoo","Ming-Hsuan Yang","Songhwai Oh"],"abstract":"We introduce a new problem of generating an image based on a small number of\nkey local patches without any geometric prior. In this work, key local patches\nare defined as informative regions of the target object or scene. This is a\nchallenging problem since it requires generating realistic images and\npredicting locations of parts at the same time. We construct adversarial\nnetworks to tackle this problem. A generator network generates a fake image as\nwell as a mask based on the encoder-decoder framework. On the other hand, a\ndiscriminator network aims to detect fake images. The network is trained with\nthree losses to consider spatial, appearance, and adversarial information. The\nspatial loss determines whether the locations of predicted parts are correct.\nInput patches are restored in the output image without much modification due to\nthe appearance loss. The adversarial loss ensures output images are realistic.\nThe proposed network is trained without supervisory signals since no labels of\nkey parts are required. Experimental results on six datasets demonstrate that\nthe proposed algorithm performs favorably on challenging objects and scenes.","url_abs":"http://arxiv.org/abs/1703.10730v2","url_pdf":"http://arxiv.org/pdf/1703.10730v2.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":"unsupervised-holistic-image-generation-from","repo_url":"https://github.com/hellbell/KeyPatchGan","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"decoder","task_name":"Decoder"},{"task_slug":"image-generation","task_name":"Image Generation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}