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 C4.8\, Level 4 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231214T094000
DTEND;TZID=Australia/Melbourne:20231214T095000
UID:siggraphasia_SIGGRAPH Asia 2023_sess148_papers_401@linklings.com
SUMMARY:HyperDreamer: Hyper-Realistic 3D Content Generation and Editing fr
 om a Single Image
DESCRIPTION:Technical Papers, TOG\n\nTong Wu and Zhibing Li (The Chinese U
 niversity of Hong Kong, Shanghai AI Laboratory); Shuai Yang (Shanghai Jiao
  Tong University, Shanghai AI Laboratory); Pan Zhang (Shanghai AI Laborato
 ry); Xingang Pan (Max Planck Institute for Informatics); Jiaqi Wang (Shang
 hai AI Laboratory); Dahua Lin (The Chinese University of Hong Kong, Shangh
 ai AI Laboratory); and Ziwei Liu (Nanyang Technological University)\n\n3D 
 content creation from a single image is a long-standing yet highly desirab
 le task. Recent advances introduce 2D diffusion priors, yielding reasonabl
 e results. However, existing methods are not hyper-realistic enough for po
 st-generation usage, as users cannot view, render and edit the resulting 3
 D content from a full range. To address these challenges, we introduce Hyp
 er-Dreamer with several key designs and appealing properties: 1) Full-rang
 e viewable: 360◦ mesh modeling with high-resolution textures enables the c
 reation of visually compelling 3D models from a full range of observation 
 points. 2) Full-range renderable: Fine-grained semantic segmentation and d
 ata-driven priors are incorporated as guidance to learn reasonable albedo,
  roughness, and specular properties of the materials, enabling semantic-aw
 are arbitrary material estimation. 3) Full-range editable: For a generated
  model or their own data, users can interactively select any region via a 
 few clicks and efficiently edit the texture with text-based guidance. Exte
 nsive experiments demonstrate the effectiveness of HyperDreamer in modelin
 g region-aware materials with high-resolution textures and enabling user-f
 riendly editing. We believe that HyperDreamer holds promise for adVancing 
 3D content creation and finding applications in various domains.\n\nRegist
 ration Category: Full Access\n\nSession Chair: Lin Lu (Shandong University
 )
URL:https://asia.siggraph.org/2023/full-program?id=papers_401&sess=sess148
END:VEVENT
END:VCALENDAR