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:20260114T163653Z
LOCATION:Meeting Room C4.11\, Level 4 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231214T144500
DTEND;TZID=Australia/Melbourne:20231214T150000
UID:siggraphasia_SIGGRAPH Asia 2023_sess130_papers_448@linklings.com
SUMMARY:MIPS-Fusion: Multi-Implicit-Submaps for Scalable and Robust Online
  Neural RGB-D Reconstruction
DESCRIPTION:Yijie Tang (National University of Defense Technology (NUDT)),
  Jiazhao Zhang (Peking University), Zhinan Yu (National University of Defe
 nse Technology (NUDT)), He Wang (Peking University), and Kai Xu (National 
 University of Defense Technology (NUDT))\n\nWe introduce MIPS-Fusion, a ro
 bust and scalable online RGB-D reconstruction method based on a novel neur
 al implicit representation -- multi-implicit-submap. Different from existi
 ng neural RGB-D reconstruction methods lacking either flexibility with a s
 ingle neural map or scalability due to extra storage of feature grids, we 
 propose a pure neural representation tackling both difficulties with a div
 ide-and-conquer design. In our method, neural submaps are incrementally al
 located alongside the scanning trajectory and efficiently learned with loc
 al neural bundle adjustments. The submaps can be refined individually in a
  back-end optimization and optimized jointly to realize submap-level loop 
 closure. Meanwhile, we propose a hybrid tracking approach combining random
 ized and gradient-based pose optimizations. For the first time, randomized
  optimization is made possible in neural tracking with several key designs
  to the learning process, enabling efficient and robust tracking even unde
 r fast camera motions. The extensive evaluation demonstrates that our meth
 od attains higher reconstruction quality than the state of the arts for la
 rge-scale scenes and under fast camera motions.\n\nRegistration Category: 
 Full Access\n\nSession Chair: Baoquan Chen (Peking University)\n\n
URL:https://asia.siggraph.org/2023/full-program?id=papers_448&sess=sess130
END:VEVENT
END:VCALENDAR