BEGIN:VCALENDAR VERSION:2.0 PRODID:Linklings LLC BEGIN:VTIMEZONE TZID:Australia/Melbourne X-LIC-LOCATION:Australia/Melbourne BEGIN:DAYLIGHT TZOFFSETFROM:+1000 TZOFFSETTO:+1100 TZNAME:AEDT DTSTART:19721003T020000 RRULE:FREQ=YEARLY;BYMONTH=4;BYDAY=1SU END:DAYLIGHT BEGIN:STANDARD DTSTART:19721003T020000 TZOFFSETFROM:+1100 TZOFFSETTO:+1000 TZNAME:AEST RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=1SU END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20240214T070247Z LOCATION:Meeting Room C4.8\, Level 4 (Convention Centre) DTSTART;TZID=Australia/Melbourne:20231214T140000 DTEND;TZID=Australia/Melbourne:20231214T141000 UID:siggraphasia_SIGGRAPH Asia 2023_sess151_papers_759@linklings.com SUMMARY:Neural Collision Fields for Triangle Primitives DESCRIPTION:Technical Papers\n\nRyan Zesch (Texas A&M University), Vismay Modi (University of Toronto), Shinjiro Sueda (Texas A&M University), and D avid Levin (University of Toronto)\n\nWe present neural collision fields a s an alternative to contact point sampling in physics simulations.\nOur ap proach is built on top of a novel smoothed integral formulation for the co ntact surface patches between two triangle meshes. By reformulating collis ions as an integral, we avoid issues of sampling common to many collision- handling algorithms. Because the resulting integral is difficult to evalua te numerically, we store its solution in an integrated neural collision fi eld --- a 6D neural field in the space of triangle pair vertex coordinates . Our network generalizes well to new triangle meshes without retraining. We demonstrate the effectiveness of our method by implementing it as a con straint in a position-based dynamics framework and show that our neural fo rmulation successfully handles collisions in practical simulations involvi ng both volumetric and thin-shell geometries.\n\nRegistration Category: Fu ll Access\n\nSession Chair: Tao Du (Tsinghua University, Shanghai Qi Zhi I nstitute) URL:https://asia.siggraph.org/2023/full-program?id=papers_759&sess=sess151 END:VEVENT END:VCALENDAR