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:20240214T070245Z
LOCATION:Meeting Room C4.8\, Level 4 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231213T152000
DTEND;TZID=Australia/Melbourne:20231213T153500
UID:siggraphasia_SIGGRAPH Asia 2023_sess145_papers_591@linklings.com
SUMMARY:Discontinuity-Aware 2D Neural Fields
DESCRIPTION:Technical Papers\n\nYash Belhe (University of California San D
 iego); Michael Gharbi, Matt Fisher, and Iliyan Georgiev (Adobe Inc.); and 
 Ravi Ramamoorthi and Tzu-Mao Li (University of California San Diego)\n\nNe
 ural image representations offer the possibility of high-fidelity, compact
  storage, and resolution-independent accuracy, providing an attractive alt
 ernative to traditional pixel and grid-based representations.  \nHowever, 
 coordinate neural networks fail to capture discontinuities present in the 
 image and tend to blur across them; we aim to address this challenge.\nFor
  many applications, such as representing a resolution-independent rendered
  image, vector graphics, diffusion curves, or solutions to partial differe
 ntial equations, we already know the locations of the discontinuities.\nWe
  take the discontinuity locations as input, represented as linear, quadrat
 ic, or cubic Bezier curves,  and construct a feature field that is only di
 scontinuous across these locations, and smooth everywhere else.\nFinally, 
 we use a shallow multi-layer perceptron to decode the features into the si
 gnal value.\nFor the feature field construction, we develop a new data str
 ucture based on a curved triangular mesh with features stored on the verti
 ces and a subset of the edges of the mesh being marked discontinuous.\nWe 
 show that our method can be used to compress a 100k^2 rendered image into 
 a 25MB file; \ncan be used as a new diffusion curve solver by combining wi
 th Monte-Carlo-based methods or directly supervised by the diffusion curve
  energy;\nor can be used for compressing 2D physics simulation data.\n\nRe
 gistration Category: Full Access\n\nSession Chair: Young J. Kim (Ewha Woma
 ns University)
URL:https://asia.siggraph.org/2023/full-program?id=papers_591&sess=sess145
END:VEVENT
END:VCALENDAR