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:20251218T030653Z
LOCATION:Meeting Room S221\, Level 2
DTSTART;TZID=Asia/Hong_Kong:20251215T153300
DTEND;TZID=Asia/Hong_Kong:20251215T154400
UID:siggraphasia_SIGGRAPH Asia 2025_sess111_papers_2228@linklings.com
SUMMARY:Neural Octahedral Field: Octahedral Prior for Simultaneous Smoothi
 ng and Sharp Edge Regularization
DESCRIPTION:Ruichen Zheng (Tsinghua University, Shenzhen University (SZU))
 ; Tao Yu (Tsinghua University); and Ruizhen Hu (Shenzhen University (SZU))
 \n\nNeural implicit representation, the parameterization of a continuous d
 istance function as a Multi-Layer Perceptron (MLP), has emerged as a promi
 sing lead in tackling surface reconstruction from unoriented point clouds.
  In the presence of noise, however, its lack of explicit neighborhood conn
 ectivity makes sharp edges identification particularly challenging, hence 
 preventing the separation of smoothing and sharpening operations, as is ac
 hievable with its discrete counterparts. In this work, we propose to tackl
 e this challenge with an auxiliary field, the \emph{octahedral field}. We 
 observe that both smoothness and sharp features in the distance field can 
 be equivalently described by the smoothness in octahedral space. Therefore
 , by aligning and smoothing an octahedral field alongside the implicit geo
 metry, our method behaves analogously to bilateral filtering, resulting in
  a smooth reconstruction while preserving sharp edges. Despite being opera
 ted purely pointwise, our method outperforms various traditional and neura
 l implicit fitting approaches across extensive experiments, and is very co
 mpetitive with methods that require normals and data priors. Code and data
  of our work are available at: https://github.com/Ankbzpx/frame-field.\n\n
 Registration Category: Full Access, Full Access Supporter\n\nSession Chair
 : Ligang Liu (University of Science and Technology of China)\n\n
END:VEVENT
END:VCALENDAR