{"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/storygan-a-sequential-conditional-gan-for","title":"StoryGAN: A Sequential Conditional GAN for Story Visualization","arxiv_id":"1812.02784","date":"2018-12-06","proceeding":"CVPR 2019 6","authors":["Yitong Li","Zhe Gan","Yelong Shen","Jingjing Liu","Yu Cheng","Yuexin Wu","Lawrence Carin","David Carlson","Jianfeng Gao"],"abstract":"We propose a new task, called Story Visualization. Given a multi-sentence\nparagraph, the story is visualized by generating a sequence of images, one for\neach sentence. In contrast to video generation, story visualization focuses\nless on the continuity in generated images (frames), but more on the global\nconsistency across dynamic scenes and characters -- a challenge that has not\nbeen addressed by any single-image or video generation methods. We therefore\npropose a new story-to-image-sequence generation model, StoryGAN, based on the\nsequential conditional GAN framework. Our model is unique in that it consists\nof a deep Context Encoder that dynamically tracks the story flow, and two\ndiscriminators at the story and image levels, to enhance the image quality and\nthe consistency of the generated sequences. To evaluate the model, we modified\nexisting datasets to create the CLEVR-SV and Pororo-SV datasets. Empirically,\nStoryGAN outperforms state-of-the-art models in image quality, contextual\nconsistency metrics, and human evaluation.","url_abs":"http://arxiv.org/abs/1812.02784v2","url_pdf":"http://arxiv.org/pdf/1812.02784v2.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":"storygan-a-sequential-conditional-gan-for","repo_url":"https://github.com/yitong91/StoryGAN","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"sentence","task_name":"Sentence"},{"task_slug":"story-visualization","task_name":"Story Visualization"},{"task_slug":"video-generation","task_name":"Video Generation"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1812.02784","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}