{"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/aligning-latent-and-image-spaces-to-connect","title":"Aligning Latent and Image Spaces to Connect the Unconnectable","arxiv_id":"2104.06954","date":"2021-04-14","proceeding":"ICCV 2021 10","authors":["Ivan Skorokhodov","Grigorii Sotnikov","Mohamed Elhoseiny"],"abstract":"In this work, we develop a method to generate infinite high-resolution images with diverse and complex content. It is based on a perfectly equivariant generator with synchronous interpolations in the image and latent spaces. Latent codes, when sampled, are positioned on the coordinate grid, and each pixel is computed from an interpolation of the nearby style codes. We modify the AdaIN mechanism to work in such a setup and train the generator in an adversarial setting to produce images positioned between any two latent vectors. At test time, this allows for generating complex and diverse infinite images and connecting any two unrelated scenes into a single arbitrarily large panorama. Apart from that, we introduce LHQ: a new dataset of \\lhqsize high-resolution nature landscapes. We test the approach on LHQ, LSUN Tower and LSUN Bridge and outperform the baselines by at least 4 times in terms of quality and diversity of the produced infinite images. The project page is located at https://universome.github.io/alis.","url_abs":"https://arxiv.org/abs/2104.06954v1","url_pdf":"https://arxiv.org/pdf/2104.06954v1.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":"aligning-latent-and-image-spaces-to-connect","repo_url":"https://github.com/universome/alis","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"diversity","task_name":"Diversity"},{"task_slug":"image-generation","task_name":"Image Generation"},{"task_slug":"infinite-image-generation","task_name":"Infinite Image Generation"}],"methods":[{"method_slug":"alis","method_name":"ALIS"},{"method_slug":"adaptive-instance-normalization","method_name":"Adaptive Instance Normalization"}],"datasets_introduced":[{"slug":"lhq","name":"LHQ","full_name":"Landscapes High-Quality"}],"methods_introduced":[{"slug":"alis","name":"ALIS","full_name":"Aligning Latent and Image Spaces"}],"results":[{"leaderboard":"/sota/infinite-image-generation-on-lhq","task":"Infinite Image Generation","dataset":"LHQ","model":"ALIS","rank_in_archive_order":1,"of":1,"metrics":{"InfinityFID":"7.8"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=2104.06954","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}