{"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/data-driven-synthesis-of-smoke-flows-with-cnn","title":"Data-Driven Synthesis of Smoke Flows with CNN-based Feature Descriptors","arxiv_id":"1705.01425","date":"2017-05-03","proceeding":null,"authors":["Mengyu Chu","Nils Thuerey"],"abstract":"We present a novel data-driven algorithm to synthesize high-resolution flow\nsimulations with reusable repositories of space-time flow data. In our work, we\nemploy a descriptor learning approach to encode the similarity between fluid\nregions with differences in resolution and numerical viscosity. We use\nconvolutional neural networks to generate the descriptors from fluid data such\nas smoke density and flow velocity. At the same time, we present a deformation\nlimiting patch advection method which allows us to robustly track deformable\nfluid regions. With the help of this patch advection, we generate stable\nspace-time data sets from detailed fluids for our repositories. We can then use\nour learned descriptors to quickly localize a suitable data set when running a\nnew simulation. This makes our approach very efficient, and resolution\nindependent. We will demonstrate with several examples that our method yields\nvolumes with very high effective resolutions, and non-dissipative small scale\ndetails that naturally integrate into the motions of the underlying flow.","url_abs":"http://arxiv.org/abs/1705.01425v2","url_pdf":"http://arxiv.org/pdf/1705.01425v2.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":"data-driven-synthesis-of-smoke-flows-with-cnn","repo_url":"https://github.com/RachelCmy/mantaPatch","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":null}],"tasks":[],"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}