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:20260114T163645Z
LOCATION:Meeting Room C4.8\, Level 4 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231215T155000
DTEND;TZID=Australia/Melbourne:20231215T160500
UID:siggraphasia_SIGGRAPH Asia 2023_sess159_papers_142@linklings.com
SUMMARY:Neural-Singular-Hessian: Implicit Neural Representation of Unorien
 ted Point Clouds by Enforcing Singular Hessian
DESCRIPTION:Zixiong Wang, Yunxiao Zhang, and Rui Xu (Shandong University);
  Fan Zhang (Shandong Technology and Business University); Peng-Shuai Wang 
 (Peking University); Shuangmin Chen (Qingdao University of Science and Tec
 hnology); Shiqing Xin (Shandong University); Wenping Wang (Texas A&M Unive
 rsity); and Changhe Tu (Shandong University)\n\nNeural implicit representa
 tion is a promising approach for reconstructing surfaces from point clouds
 . Existing methods combine various regularization terms to enforce the lea
 rned neural function to possess the properties of a SDF, such as the Eikon
 al term and Laplacian energy term. However, when the input is an unoriente
 d point cloud of poor quality, inferring the actual topology and geometry 
 of the underlying surface becomes challenging. In accordance with Differen
 tial Geometry, the Hessian of the SDF is singular for any point located wi
 thin the differential thin-shell space surrounding the surface. Based on t
 his fact, our approach enforces the Hessian of the SDF to have a zero dete
 rminant for points in close proximity to the input point cloud, rather tha
 n using a smoothness term. This technique quickly eliminates critical poin
 ts of the SDF near the surface, producing a coarse but faithful shape. By 
 annealing the weight of the critical-point elimination term, our approach 
 can ultimately produce a high-fidelity reconstruction result. The validity
  of this approach can be established through Morse theory. Our approach ha
 s demonstrated better expressiveness in recovering details from unoriented
  point clouds while effectively suppressing ghost geometry, as evidenced b
 y extensive experimental results in comparison to existing fitting-based m
 ethods.\n\nRegistration Category: Full Access\n\nSession Chair: Fei Hou (I
 nstitute of Software, Chinese Academy of Sciences; University of Chinese A
 cademy of Sciences)\n\n
URL:https://asia.siggraph.org/2023/full-program?id=papers_142&sess=sess159
END:VEVENT
END:VCALENDAR