{"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/nuwa-infinity-autoregressive-over","title":"NUWA-Infinity: Autoregressive over Autoregressive Generation for Infinite Visual Synthesis","arxiv_id":"2207.09814","date":"2022-07-20","proceeding":null,"authors":["Chenfei Wu","Jian Liang","Xiaowei Hu","Zhe Gan","JianFeng Wang","Lijuan Wang","Zicheng Liu","Yuejian Fang","Nan Duan"],"abstract":"In this paper, we present NUWA-Infinity, a generative model for infinite visual synthesis, which is defined as the task of generating arbitrarily-sized high-resolution images or long-duration videos. An autoregressive over autoregressive generation mechanism is proposed to deal with this variable-size generation task, where a global patch-level autoregressive model considers the dependencies between patches, and a local token-level autoregressive model considers dependencies between visual tokens within each patch. A Nearby Context Pool (NCP) is introduced to cache-related patches already generated as the context for the current patch being generated, which can significantly save computation costs without sacrificing patch-level dependency modeling. An Arbitrary Direction Controller (ADC) is used to decide suitable generation orders for different visual synthesis tasks and learn order-aware positional embeddings. Compared to DALL-E, Imagen and Parti, NUWA-Infinity can generate high-resolution images with arbitrary sizes and support long-duration video generation additionally. Compared to NUWA, which also covers images and videos, NUWA-Infinity has superior visual synthesis capabilities in terms of resolution and variable-size generation. The GitHub link is https://github.com/microsoft/NUWA. The homepage link is https://nuwa-infinity.microsoft.com.","url_abs":"https://arxiv.org/abs/2207.09814v2","url_pdf":"https://arxiv.org/pdf/2207.09814v2.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":"nuwa-infinity-autoregressive-over","repo_url":"https://github.com/microsoft/nuwa","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"image-outpainting","task_name":"Image Outpainting"},{"task_slug":"text-to-image-generation","task_name":"Text-to-Image Generation"},{"task_slug":"video-generation","task_name":"Video Generation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-outpainting-on-lhqc","task":"Image Outpainting","dataset":"LHQC","model":"NUWA-Infinity w/o text","rank_in_archive_order":1,"of":4,"metrics":{"Block-FID  (Right Extend)":"6.43","Block-FID (Down Extend)":"11.47","Block-FID (Left Extend)":"6.71","Block-FID (Up Extend)":"8.03"},"uses_additional_data":false},{"leaderboard":"/sota/image-outpainting-on-lhqc","task":"Image Outpainting","dataset":"LHQC","model":"NUWA-Infinity","rank_in_archive_order":2,"of":4,"metrics":{"Block-FID  (Right Extend)":"6.45","Block-FID (Down Extend)":"9.84","Block-FID (Left Extend)":"6.72","Block-FID (Up Extend)":"7.43"},"uses_additional_data":false},{"leaderboard":"/sota/text-to-image-generation-on-lhqc","task":"Text-to-Image Generation","dataset":"LHQC","model":"NUWA-Infinity","rank_in_archive_order":1,"of":3,"metrics":{"Block-FID":"9.71"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2207.09814","atlas_url":"https://app.syntology.ai/?focus=2207.09814","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}