BEGIN:VCALENDAR VERSION:2.0 PRODID:Linklings LLC BEGIN:VTIMEZONE TZID:Asia/Seoul X-LIC-LOCATION:Asia/Seoul BEGIN:STANDARD TZOFFSETFROM:+0900 TZOFFSETTO:+0900 TZNAME:KST DTSTART:18871231T000000 DTSTART:19881009T020000 END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20230103T035312Z LOCATION:Room 325-AB\, Level 3\, West Wing DTSTART;TZID=Asia/Seoul:20221209T090000 DTEND;TZID=Asia/Seoul:20221209T103000 UID:siggraphasia_SIGGRAPH Asia 2022_sess176_papers_522@linklings.com SUMMARY:Efficient Neural Style Transfer for Volumetric Simulations DESCRIPTION:Technical Communications, Technical Papers, TOG\n\nEfficient N eural Style Transfer for Volumetric Simulations\n\nAurand, Ortiz, Nauer, C . Azevedo\n\nArtistically controlling fluids has always been a challenging task. Recently, volumetric Neural Style Transfer (NST) techniques have be en used to artistically manipulate smoke simulation data with 2D images. I n this work, we revisit previous volumetric NST techniques for smoke, prop osing a suite of upgrades that enable stylizations that are significantly faster, simpler, more controllable and less prone to artifacts. \nMoreover , the energy minimization solved by previous methods is camera dependent. To avoid that, a computationally expensive iterative optimization perform ed for multiple views sampled around the original simulation is needed, wh ich can take up to several minutes per frame. We propose a simple feed-for ward neural network architecture that is able to infer view-independent st ylizations that are three orders of the magnitude faster than its optimiza tion-based counterpart.\n\nRegistration Category: FULL ACCESS, ON-DEMAND A CCESS\n\nLanguage: ENGLISH\n\nFormat: IN-PERSON, ON-DEMAND URL:https://sa2022.siggraph.org/en/full-program/?id=papers_522&sess=sess17 6 END:VEVENT END:VCALENDAR