BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Asia/Hong_Kong
X-LIC-LOCATION:Asia/Hong_Kong
BEGIN:STANDARD
TZOFFSETFROM:+0800
TZOFFSETTO:+0800
TZNAME:HKT
DTSTART:19911015T033000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20251218T030656Z
LOCATION:Meeting Room S423+S424\, Level 4
DTSTART;TZID=Asia/Hong_Kong:20251217T111200
DTEND;TZID=Asia/Hong_Kong:20251217T112300
UID:siggraphasia_SIGGRAPH Asia 2025_sess135_papers_1483@linklings.com
SUMMARY:PanoDreamer: Optimization-Based Single Image to 360 3D Scene With 
 Diffusion
DESCRIPTION:Avinash Paliwal (Texas A&M University, Morphic Inc); Xilong Zh
 ou (Max Planck Institute for Informatics); Andrii Tsarov (Leia Inc.); and 
 Nima Kalantari (Texas A&M University)\n\nIn this paper, we present PanoDre
 amer, a novel method for producing a coherent 360° 3D scene from a single 
 input image. Unlike existing methods that generate the scene sequentially,
  we frame the problem as single-image panorama and depth estimation. Once 
 the coherent panoramic image and its corresponding depth are obtained, the
  scene can be reconstructed by inpainting the small occluded regions and p
 rojecting them into 3D space. Our key contribution is formulating single-i
 mage panorama and depth estimation as two optimization tasks and introduci
 ng alternating minimization strategies to effectively solve their objectiv
 es. We demonstrate that our approach outperforms existing techniques in si
 ngle-image 360° 3D scene reconstruction in terms of consistency and overal
 l quality.\n\nRegistration Category: Full Access, Full Access Supporter\n\
 nSession Chair: Or Patashnik (Tel Aviv University, Snap Research)\n\n
END:VEVENT
END:VCALENDAR