{"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/object-driven-text-to-image-synthesis-via","title":"Object-driven Text-to-Image Synthesis via Adversarial Training","arxiv_id":"1902.10740","date":"2019-02-27","proceeding":"CVPR 2019 6","authors":["Wenbo Li","Pengchuan Zhang","Lei Zhang","Qiuyuan Huang","Xiaodong He","Siwei Lyu","Jianfeng Gao"],"abstract":"In this paper, we propose Object-driven Attentive Generative Adversarial\nNewtorks (Obj-GANs) that allow object-centered text-to-image synthesis for\ncomplex scenes. Following the two-step (layout-image) generation process, a\nnovel object-driven attentive image generator is proposed to synthesize salient\nobjects by paying attention to the most relevant words in the text description\nand the pre-generated semantic layout. In addition, a new Fast R-CNN based\nobject-wise discriminator is proposed to provide rich object-wise\ndiscrimination signals on whether the synthesized object matches the text\ndescription and the pre-generated layout. The proposed Obj-GAN significantly\noutperforms the previous state of the art in various metrics on the large-scale\nCOCO benchmark, increasing the Inception score by 27% and decreasing the FID\nscore by 11%. A thorough comparison between the traditional grid attention and\nthe new object-driven attention is provided through analyzing their mechanisms\nand visualizing their attention layers, showing insights of how the proposed\nmodel generates complex scenes in high quality.","url_abs":"http://arxiv.org/abs/1902.10740v1","url_pdf":"http://arxiv.org/pdf/1902.10740v1.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":"object-driven-text-to-image-synthesis-via","repo_url":"https://github.com/jamesli1618/Obj-GAN","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"image-generation","task_name":"Image Generation"},{"task_slug":"object","task_name":"Object"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"fast-r-cnn","method_name":"Fast R-CNN"},{"method_slug":"roipool","method_name":"RoIPool"},{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1902.10740","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}