BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Asia/Seoul
X-LIC-LOCATION:Asia/Seoul
BEGIN:STANDARD
TZOFFSETFROM:+0900
TZOFFSETTO:+0900
TZNAME:KST
DTSTART:18871231T000000
DTSTART:19881009T020000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20230103T035311Z
LOCATION:Room 325-AB\, Level 3\, West Wing
DTSTART;TZID=Asia/Seoul:20221208T140000
DTEND;TZID=Asia/Seoul:20221208T153000
UID:siggraphasia_SIGGRAPH Asia 2022_sess169_papers_381@linklings.com
SUMMARY:Neural Parameterization for Dynamic Human Head Editing
DESCRIPTION:Technical Communications, Technical Papers\n\nNeural Parameter
 ization for Dynamic Human Head Editing\n\nMa, Li, Liao, Wang, Zhang...\n\n
 Implicit radiance functions emerged as a powerful scene representation for
  reconstructing and rendering photo-realistic views of a 3D scene. These r
 epresentations, however, suffer from poor editability. On the other hand, 
 explicit representations such as polygonal meshes allow easy editing, but 
 are not as suitable for reconstructing accurate details in dynamic human h
 eads, such as fine facial features, hair, and eyes. In this work, we prese
 nt Neural Parameterization (NeP), a hybrid representation that provides th
 e advantages of both implicit and explicit methods. NeP is capable of phot
 o-realistic rendering while allowing fine-grained editing of the scene geo
 metry and appearance. We first disentangle the geometry and appearance by 
 parameterizing the 3D geometry into 2D texture space. We enable geometric 
 editability by introducing an explicit linear deformation blending layer. 
 The deformation is controlled by a set of sparse key points which can be e
 xplicitly and intuitively displaced to edit the geometry. For appearance, 
 we develop a hybrid 2D texture consisting of an explicit texture map for e
 asy editing and implicit view and time-dependent residuals to model tempor
 al and view variations. We compare our method to several reconstruction an
 d editing baselines. The results show that the NeP achieves almost the sam
 e level of rendering accuracy while maintaining high editability.\n\nRegis
 tration Category: FULL ACCESS, ON-DEMAND ACCESS\n\nLanguage: ENGLISH\n\nFo
 rmat: IN-PERSON, ON-DEMAND
URL:https://sa2022.siggraph.org/en/full-program/?id=papers_381&sess=sess16
 9
END:VEVENT
END:VCALENDAR