{"673338":{"#nid":"673338","#data":{"type":"news","title":"Researchers Reach New AI Benchmark for Computer Graphics","body":[{"value":"\u003Cp\u003EComputer graphic simulations can represent natural phenomena such as tornados, underwater, vortices, and liquid foams more accurately thanks to an advancement in creating artificial intelligence (AI) neural networks.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EWorking with a multi-institutional team of researchers, Georgia Tech Assistant Professor Bo Zhu combined computer graphic simulations with machine learning models to create enhanced simulations of known phenomena. The new benchmark could lead to researchers constructing representations of other phenomena that have yet to be simulated.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EZhu co-authored the paper\u0026nbsp;\u003Cem\u003EFluid Simulation on Neural Flow Maps\u003C\/em\u003E. The Association for Computing Machinery\u2019s Special Interest Group in Computer Graphics and Interactive Technology\u0026nbsp;\u003Ca href=\u0022https:\/\/www.siggraph.org\/\u0022\u003E(SIGGRAPH)\u003C\/a\u003E\u0026nbsp;gave it a best paper award in December at the SIGGRAPH Asia conference in Sydney, Australia.\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003EThe authors say the advancement could be as significant to computer graphic simulations as the introduction of\u0026nbsp;\u003Ca href=\u0022https:\/\/www.matthewtancik.com\/nerf\u0022\u003Eneural radiance fields\u003C\/a\u003E\u0026nbsp;(NeRFs) was to computer vision in 2020. Introduced by researchers at the University of California-Berkley, University of California-San Diego, and Google, NeRFs are neural networks that easily convert 2D images into 3D navigable scenes.\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003ENeRFs have become a benchmark among computer vision researchers. Zhu and his collaborators hope their creation, neural flow maps, can do the same for simulation researchers in computer graphics.\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u201cA natural question to ask is, can AI fundamentally overcome the traditional method\u2019s shortcomings and bring generational leaps to simulation as it has done to natural language processing and computer vision?\u201d Zhu said. \u201cSimulation accuracy has been a significant challenge to computer graphics researchers. No existing work has combined AI with physics to yield high-end simulation results that outperform traditional schemes in accuracy.\u201d\u003C\/p\u003E\r\n\r\n\u003Cp\u003EIn computer graphics, simulation pipelines are the equivalent of neural networks and allow simulations to take shape. They are traditionally constructed through mathematical equations and numerical schemes.\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003EZhu said researchers have tried to design simulation pipelines with neural representations to construct more robust simulations. However, efforts to achieve higher physical accuracy have fallen short.\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003EZhu attributes the problem to the pipelines\u2019 incapability of matching the capacities of AI algorithms within the structures of traditional simulation pipelines. To solve the problem and allow machine learning to have influence, Zhu and his collaborators proposed a new framework that redesigns the simulation pipeline.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EThey named these new pipelines neural flow maps. The maps use machine learning models to store spatiotemporal data more efficiently. The researchers then align these models with their mathematical framework to achieve a higher accuracy than previous pipeline simulations.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EZhu said he does not believe machine learning should be used to replace traditional numerical equations. Rather, they should complement them to unlock new advantageous paradigms.\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u201cInstead of trying to deploy modern AI techniques to replace components inside traditional pipelines, we co-designed the simulation algorithm and machine learning technique in tandem,\u201d Zhu said.\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u201cNumerical methods are not optimal because of their limited computational capacity. Recent AI-driven capacities have uplifted many of these limitations. Our task is redesigning existing simulation pipelines to take full advantage of these new AI capacities.\u201d\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003EIn the paper, the authors state the once unattainable algorithmic designs could unlock new research possibilities in computer graphics.\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003ENeural flow maps offer \u201ca new perspective on the incorporation of machine learning in numerical simulation research for computer graphics and computational sciences alike,\u201d the paper states.\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u201cThe success of Neural Flow Maps is inspiring for how physics and machine learning are best combined,\u201d Zhu added.\u003C\/p\u003E\r\n","summary":"","format":"limited_html"}],"field_subtitle":"","field_summary":[{"value":"\u003Cp\u003EAn interactive computing Assistant Professor Bo Zhu has earned a best paper award from a leading computer graphics and interactivity conference for his work combining computer graphic simulations with machine learning models to create enhanced simulations of known phenomena.\u003C\/p\u003E\r\n","format":"limited_html"}],"field_summary_sentence":[{"value":"An interactive computing faculty member has earned a best paper award from a leading computer graphics and interactivity conference."}],"uid":"32045","created_gmt":"2024-03-05 17:47:24","changed_gmt":"2024-03-05 19:05:10","author":"Ben Snedeker","boilerplate_text":"","field_publication":"","field_article_url":"","dateline":{"date":"2024-03-05T00:00:00-05:00","iso_date":"2024-03-05T00:00:00-05:00","tz":"America\/New_York"},"extras":[],"hg_media":{"673307":{"id":"673307","type":"image","title":"Bo Zho is an assistant professor in Georgia Tech\u0027s School of Interactive Computing","body":"\u003Cp\u003EGeorgia Tech Assistant Professor Bo Zhu worked on a multi-institutional team to develop a new AI benchmark for computer graphics. Photo by Eli Burakian\/Dartmouth College.\u003C\/p\u003E\r\n","created":"1709660699","gmt_created":"2024-03-05 17:44:59","changed":"1709660646","gmt_changed":"2024-03-05 17:44:06","alt":"Bo Zho is an assistant professor in Georgia Tech\u0027s School of Interactive Computing","file":{"fid":"256680","name":"Bo Zho-march24.jpeg","image_path":"\/sites\/default\/files\/2024\/03\/05\/Bo%20Zho-march24.jpeg","image_full_path":"http:\/\/hg.gatech.edu\/\/sites\/default\/files\/2024\/03\/05\/Bo%20Zho-march24.jpeg","mime":"image\/jpeg","size":74659,"path_740":"http:\/\/hg.gatech.edu\/sites\/default\/files\/styles\/740xx_scale\/public\/2024\/03\/05\/Bo%20Zho-march24.jpeg?itok=VUC8_5pG"}}},"media_ids":["673307"],"related_links":[{"url":"https:\/\/youtu.be\/VPDsOcG_zW0","title":"What are Neural Flow Maps?"}],"groups":[{"id":"47223","name":"College of Computing"}],"categories":[{"id":"153","name":"Computer Science\/Information Technology and Security"}],"keywords":[{"id":"10199","name":"Daily Digest"},{"id":"187915","name":"go-researchnews"},{"id":"192863","name":"go-ai"}],"core_research_areas":[{"id":"39501","name":"People and Technology"},{"id":"39541","name":"Systems"}],"news_room_topics":[],"event_categories":[],"invited_audience":[],"affiliations":[],"classification":[],"areas_of_expertise":[],"news_and_recent_appearances":[],"phone":[],"contact":[{"value":"\u003Cp\u003ENathan Deen, Communications Officer\u003C\/p\u003E\r\n\r\n\u003Cp\u003EGeorgia Tech School of Interactive Computing\u003C\/p\u003E\r\n\r\n\u003Cp\u003Enathan.deen@cc.gatech.edu\u003C\/p\u003E\r\n","format":"limited_html"}],"email":[],"slides":[],"orientation":[],"userdata":""}}}