BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Asia/Tokyo
X-LIC-LOCATION:Asia/Tokyo
BEGIN:STANDARD
TZOFFSETFROM:+0900
TZOFFSETTO:+0900
TZNAME:JST
DTSTART:18871231T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260817T171533Z
LOCATION:Hall B5 (1)\, B Block\, Level 5
DTSTART;TZID=Asia/Tokyo:20241204T111900
DTEND;TZID=Asia/Tokyo:20241204T113100
UID:siggraphasia_SIGGRAPH Asia 2024_sess112_papers_241@linklings.com
SUMMARY:PVP-Recon: Progressive View Planning via Warping Consistency for S
 parse-View Surface Reconstruction
DESCRIPTION:Sheng Ye, Yuze He, Matthieu Lin, Jenny Sheng, and Ruoyu Fan (T
 singhua University); Yiheng Han (Beijing University of Technology); Yubin 
 Hu (Tsinghua University); Ran Yi (Shanghai Jiao Tong University); Yu-Hui W
 en (Beijing Jiaotong University); Yong-Jin Liu (Tsinghua University); and 
 Wenping Wang (Texas A&M University)\n\nNeural implicit representations hav
 e revolutionized dense multi-view surface reconstruction, yet their perfor
 mance significantly diminishes with sparse input views. A few pioneering w
 orks have sought to tackle the challenge of sparse-view reconstruction by 
 leveraging additional geometric priors or multi-scene generalizability. Ho
 wever, they are still hindered by the imperfect choice of input views, usi
 ng images under empirically determined viewpoints to provide considerable 
 overlap. We propose PVP-Recon, a novel and effective sparse-view surface r
 econstruction method that progressively plans the next best views to form 
 an optimal set of sparse viewpoints for image capturing. PVP-Recon starts 
 initial surface reconstruction with as few as 3 views and progressively ad
 ds new views which are determined based on a novel warping score that refl
 ects the information gain of each newly added view. This progressive view 
 planning progress is interleaved with a neural SDF-based reconstruction mo
 dule that utilizes multi-resolution hash features, enhanced by a progressi
 ve training scheme and a directional Hessian loss. Quantitative and qualit
 ative experiments on three benchmark datasets show that our framework achi
 eves high-quality reconstruction with a constrained input budget and outpe
 rforms existing baselines.\n\nRegistration Category: Full Access, Full Acc
 ess Supporter\n\nLanguage Format: English Language\n\nSession Chair: Micha
 el Wimmer (TU Wien, Technische Universität Wien (TU Wien))\n\n
URL:https://asia.siggraph.org/2024/program/?id=papers_241&sess=sess112
END:VEVENT
END:VCALENDAR