{"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/learning-to-navigate-image-manifolds-induced","title":"Learning to navigate image manifolds induced by generative adversarial networks for unsupervised video generation","arxiv_id":"1901.11384","date":"2019-01-23","proceeding":null,"authors":["Isabela Albuquerque","João Monteiro","Tiago H. Falk"],"abstract":"In this work, we introduce a two-step framework for generative modeling of\ntemporal data. Specifically, the generative adversarial networks (GANs) setting\nis employed to generate synthetic scenes of moving objects. To do so, we\npropose a two-step training scheme within which: a generator of static frames\nis trained first. Afterwards, a recurrent model is trained with the goal of\nproviding a sequence of inputs to the previously trained frames generator, thus\nyielding scenes which look natural. The adversarial setting is employed in both\ntraining steps. However, with the aim of avoiding known training instabilities\nin GANs, a multiple discriminator approach is used to train both models.\nResults in the studied video dataset indicate that, by employing such an\napproach, the recurrent part is able to learn how to coherently navigate the\nimage manifold induced by the frames generator, thus yielding more\nnatural-looking scenes.","url_abs":"http://arxiv.org/abs/1901.11384v1","url_pdf":"http://arxiv.org/pdf/1901.11384v1.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":"learning-to-navigate-image-manifolds-induced","repo_url":"https://github.com/belaalb/frameGAN","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"navigate","task_name":"Navigate"},{"task_slug":"video-generation","task_name":"Video Generation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}