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:20231214T115000
DTEND;TZID=Australia/Melbourne:20231214T120000
UID:siggraphasia_SIGGRAPH Asia 2023_sess150_papers_535@linklings.com
SUMMARY:Explorable Mesh Deformation Subspaces from Unstructured 3D Generat
 ive Models
DESCRIPTION:Technical Papers\n\nArman Maesumi (Brown University); Paul Gue
 rrero, Vladimir Kim, and Matthew Fisher (Adobe Inc.); Siddhartha Chaudhuri
  (Adobe Inc.; Indian Institute of Technology (IIT), Bombay); Noam Aigerman
  (Adobe Inc.); and Daniel Ritchie (Brown University)\n\nExploring variatio
 ns of 3D shapes is a time-consuming process in traditional 3D modeling too
 ls. Deep generative models of 3D shapes often feature continuous latent sp
 aces that can, in principle, be used to explore potential variations start
 ing from a set of input shapes; in practice, doing so can be problematic--
 -latent spaces are high dimensional and hard to visualize, contain shapes 
 that are not relevant to the input shapes, and linear paths through them o
 ften lead to sub-optimal shape transitions. Furthermore, one would ideally
  be able to explore variations in the original high-quality meshes used to
  train the generative model, not its lower-quality output geometry. In thi
 s paper, we present a method to explore variations among a given set of la
 ndmark shapes by constructing a mapping from an easily-navigable 2d explor
 ation space to a subspace of a pre-trained generative model. We first desc
 ribe how to find a mapping that spans the set of input landmark shapes and
  exhibits smooth variations between them. We then show how to turn the var
 iations in this subspace into deformation fields, to transfer those variat
 ions to high-quality meshes for the landmark shapes. Our results show that
  our method can produce visually-pleasing and easily-navigable 2D explorat
 ion spaces for several different shape categories, especially as compared 
 to prior work on learning deformation spaces for 3D shapes.\n\nRegistratio
 n Category: Full Access\n\nSession Chair: Peng-Shuai Wang (Peking Universi
 ty)
URL:https://asia.siggraph.org/2023/full-program?id=papers_535&sess=sess150
END:VEVENT
END:VCALENDAR