BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Australia/Brisbane
X-LIC-LOCATION:Australia/Brisbane
BEGIN:STANDARD
DTSTART:19920301T030000
TZOFFSETFROM:+1100
TZOFFSETTO:+1000
TZNAME:AEST
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20191203T080637Z
LOCATION:Great Hall 1&2
DTSTART;TZID=Australia/Brisbane:20191117T195700
DTEND;TZID=Australia/Brisbane:20191117T195800
UID:siggraphasia_SIGGRAPH Asia 2019_sess222_papers_240@linklings.com
SUMMARY:StructureNet: Hierarchical Graph Networks for 3D Shape Generation
DESCRIPTION:Technical Papers Fast-Forward \n\nStructureNet: Hierarchical G
 raph Networks for 3D Shape Generation\n\nMo, Guerrero, Yi, Su, Wonka...\n\
 nThe ability to generate novel, diverse, and realistic 3D shapes along wit
 h associated part semantics and structure is central to many applications 
 requiring high-quality 3D assets or large volumes of realistic training da
 ta. A key challenge towards this goal is how to accommodate diverse shape 
 variations, including both continuous deformations of parts as well as str
 uctural or discrete alterations which add to, remove from, or modify the s
 hape constituents and compositional structure. Such object structure can t
 ypically be organized into a hierarchy of constituent object parts and rel
 ationships, represented as a hierarchy of n-ary graphs. We introduce Struc
 tureNet, a hierarchical graph network which (i)~can directly encode shapes
  represented as such n-ary graphs, (ii)~can be robustly trained on large a
 nd complex shape families, and (iii)~be used to generate a great diversity
  of realistic structured shape geometries. Technically, we accomplish this
  by drawing inspiration from recent advances in graph neural networks to p
 ropose an order-invariant encoding of n-ary graphs, considering jointly bo
 th part geometry and inter-part relations during network training. We exte
 nsively evaluate the quality of the learned latent spaces for various shap
 e families and show significant advantages over baseline and competing met
 hods. The learned latent spaces enable several structure-aware geometry pr
 ocessing applications, including shape generation and interpolation, shape
  editing, or shape structure discovery directly from un-annotated images, 
 point clouds, or partial scans.\n\nRegistration Category: Platinum Pass, F
 ull Conference Pass, Full Conference One-Day Pass, Basic Conference Pass, 
 Student One-Day Pass, Exhibitor Pass
URL:https://sa2019.conference-program.com/presentation?id=papers_240&sess=
 sess222
END:VEVENT
END:VCALENDAR