{"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/semantic-image-synthesis-via-adversarial","title":"Semantic Image Synthesis via Adversarial Learning","arxiv_id":"1707.06873","date":"2017-07-21","proceeding":"ICCV 2017 10","authors":["Hao Dong","Simiao Yu","Chao Wu","Yike Guo"],"abstract":"In this paper, we propose a way of synthesizing realistic images directly\nwith natural language description, which has many useful applications, e.g.\nintelligent image manipulation. We attempt to accomplish such synthesis: given\na source image and a target text description, our model synthesizes images to\nmeet two requirements: 1) being realistic while matching the target text\ndescription; 2) maintaining other image features that are irrelevant to the\ntext description. The model should be able to disentangle the semantic\ninformation from the two modalities (image and text), and generate new images\nfrom the combined semantics. To achieve this, we proposed an end-to-end neural\narchitecture that leverages adversarial learning to automatically learn\nimplicit loss functions, which are optimized to fulfill the aforementioned two\nrequirements. We have evaluated our model by conducting experiments on\nCaltech-200 bird dataset and Oxford-102 flower dataset, and have demonstrated\nthat our model is capable of synthesizing realistic images that match the given\ndescriptions, while still maintain other features of original images.","url_abs":"http://arxiv.org/abs/1707.06873v1","url_pdf":"http://arxiv.org/pdf/1707.06873v1.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":"semantic-image-synthesis-via-adversarial","repo_url":"https://github.com/vtddggg/BilinearGAN_for_LBIE","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"semantic-image-synthesis-via-adversarial","repo_url":"https://github.com/woozzu/dong_iccv_2017","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"image-generation","task_name":"Image Generation"},{"task_slug":"image-manipulation","task_name":"Image Manipulation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1707.06873","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}