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:20240214T070244Z
LOCATION:Meeting Room C4.11\, Level 4 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231213T142500
DTEND;TZID=Australia/Melbourne:20231213T143500
UID:siggraphasia_SIGGRAPH Asia 2023_sess128_papers_532@linklings.com
SUMMARY:DreamEditor: Text-Driven 3D Scene Editing with Neural Fields
DESCRIPTION:Technical Papers\n\nJingyu Zhuang (Sun Yat-sen University); Ch
 en Wang (University of Pennsylvania, Tsinghua University); Liang Lin (Sun 
 Yat-sen University); Lingjie Liu (University of Pennsylvania); and Guanbin
  Li (Sun Yat-sen University)\n\nNeural fields have achieved impressive adv
 ancements in view synthesis and scene reconstruction. However, editing the
 se neural fields remains challenging due to the implicit encoding of geome
 try and texture information. In this paper, we propose DreamEditor, a nove
 l framework that enables users to perform controlled editing of neural fie
 lds using text prompts. By representing scenes as mesh-based neural fields
 , DreamEditor allows localized editing within specific regions. DreamEdito
 r utilizes the text encoder of a pretrained text-to-Image diffusion model 
 to automatically identify the regions to be edited based on the semantics 
 of the text prompts. Subsequently, DreamEditor optimizes the edit region a
 nd aligns its geometry and texture with the text prompts through score dis
 tillation sampling. Extensive experiments have demonstrated that DreamEdit
 or can accurately edit neural fields of real-world scenes according to the
  given text prompts while ensuring consistency in irrelevant areas. DreamE
 ditor generates highly realistic textures and geometry, significantly surp
 assing previous works in both quantitative and qualitative evaluations.\n\
 nRegistration Category: Full Access\n\nSession Chair: Jianfei Cai (Monash 
 University)
URL:https://asia.siggraph.org/2023/full-program?id=papers_532&sess=sess128
END:VEVENT
END:VCALENDAR