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:20240214T070246Z
LOCATION:Meeting Room C4.11\, Level 4 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231214T095000
DTEND;TZID=Australia/Melbourne:20231214T100500
UID:siggraphasia_SIGGRAPH Asia 2023_sess124_tog_107@linklings.com
SUMMARY:NeRFFaceLighting: Implicit and Disentangled Face Lighting Represen
 tation Leveraging Generative Prior in Neural Radiance Fields
DESCRIPTION:Technical Papers, TOG\n\nKaiwen Jiang (Institute of Computing 
 Technology, Chinese Academy of Sciences; Beijing Jiaotong University); Shu
 -Yu Chen (Institute of Computing Technology, Chinese Academy of Sciences);
  Hongbo Fu (School of Creative Media, City University of Hong Kong); and L
 in Gao (Institute of Computing Technology, Chinese Academy of Sciences; Un
 iversity of Chinese Academy of Sciences)\n\n3D-aware portrait lighting con
 trol is an emerging and promising domain, thanks to the recent advance of 
 generative adversarial networks and neural radiance fields. Existing solut
 ions typically try to decouple the lighting from the geometry and appearan
 ce for disentangled control with an explicit lighting representation (e.g.
 , Lambertian or Phong). However, they either are limited to a constrained 
 lighting condition (e.g., directional light) or demand a tricky-to-fetch d
 ataset as supervision for the intrinsic compositions (e.g., the albedo). W
 e propose NeRFFaceLighting to explore an implicit representation for portr
 ait lighting based on the pretrained tri-plane representation to address t
 he above limitations. We approach this disentangled lighting-control probl
 em by distilling the shading from the original fused representation of bot
 h appearance and lighting (i.e., one tri-plane) to their disentangled repr
 esentations (i.e., two tri-planes) with the conditional discriminator to s
 upervise the lighting effects. We further carefully design the regularizat
 ion to reduce the ambiguity of such decomposition and enhance the ability 
 of generalization to unseen lighting conditions. Moreover, our method can 
 be extended to enable 3D-aware real portrait relighting. Through extensive
  quantitative and qualitative evaluations, we demonstrate the superior 3D-
 aware lighting control ability of our model compared to alternative and ex
 isting solutions.\n\nRegistration Category: Full Access\n\nSession Chair: 
 Lin Gao (University of Chinese Academy of Sciences)
URL:https://asia.siggraph.org/2023/full-program?id=tog_107&sess=sess124
END:VEVENT
END:VCALENDAR