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:20260114T163707Z
LOCATION:Meeting Room C4.8\, Level 4 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231214T121500
DTEND;TZID=Australia/Melbourne:20231214T123000
UID:siggraphasia_SIGGRAPH Asia 2023_sess150_papers_491@linklings.com
SUMMARY:GeoLatent: A Geometric Approach to Latent Space Design for Deforma
 ble Shape Generators
DESCRIPTION:Haitao Yang, Bo Sun, Liyan Chen, Amy Pavel, and Qixing Huang (
 University of Texas at Austin)\n\nWe study how to optimize the latent spac
 e of neural shape generators that map latent codes to 3D deformable shapes
 . The key focus is to look at a deformable shape generator from a differen
 tial geometry perspective. We define a Riemannian metric based on as-rigid
 -as-possible and as-conformal-as-possible deformation energies. Under this
  metric, we study two desired properties of the latent space: 1) straight-
 line interpolations in latent codes follow geodesic curves; 2) latent code
 s disentanglement pose and shape variations at different scales. Strictly 
 enforcing the geometric interpolation property, however, only applies if t
 he metric matrix is a constant. We show how to achieve this property appro
 ximately by enforcing that geodesic extrapolations are axis-aligned, i.e.,
  extrapolations along coordinate axis follow geodesic curves. In addition,
  we introduce a novel approach that decouples pose and shape variations vi
 a generalized eigendecomposition. We also study efficient regularization t
 erms for learning deformable shape generators, e.g., that promote smooth i
 nterpolations. Experimental results on benchmark datasets show that our ap
 proach leads to interpretable latent codes, improves the generalizability 
 of synthetic shapes, and enhances performance in geodesic interpolation, g
 eodesic shooting, and parallel translation applications.\n\nRegistration C
 ategory: Full Access\n\nSession Chair: Peng-Shuai Wang (Peking University)
 \n\n
URL:https://asia.siggraph.org/2023/full-program?id=papers_491&sess=sess150
END:VEVENT
END:VCALENDAR