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:20240214T070250Z
LOCATION:Meeting Room C4.11\, Level 4 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231215T133300
DTEND;TZID=Australia/Melbourne:20231215T134800
UID:siggraphasia_SIGGRAPH Asia 2023_sess137_papers_231@linklings.com
SUMMARY:EMS: 3D Eyebrow Modeling from Single-view Images
DESCRIPTION:Technical Communications, Technical Papers\n\nChenghong Li, Le
 yang Jin, and Yujian Zheng (The Chinese University of Hong Kong, Shenzhen)
 ; Yizhou Yu (The University of Hong Kong); and Xiaoguang Han (The Chinese 
 University of Hong Kong, Shenzhen)\n\nEyebrows play a critical role in fac
 ial expression and appearance. Although the 3D digitization of faces is we
 ll explored, less attention has been drawn to 3D eyebrow modeling. In this
  work, we propose EMS, the first learning-based framework for single-view 
 3D eyebrow reconstruction. Following the methods of scalp hair reconstruct
 ion, we also represent the eyebrow as a set of fiber curves and convert th
 e reconstruction to fibers growing problem. Three modules are then careful
 ly designed: RootFinder firstly localizes the fiber root positions which i
 ndicates where to grow; OriPredictor predicts an orientation filed in the 
 3D space to guide the growing of fibers; FiberEnder is designed to determi
 ne when to stop the growth of each fiber. Our OriPredictor is directly bor
 rowing the method used in hair reconstruction. Considering the differences
  between hair and eyebrows, both RootFinder and FiberEnder are newly propo
 sed. Specifically, to cope with the challenge that the root location is se
 verely occluded, we formulate root localization as a density map estimatio
 n task. Given the predicted density map, a density-based clustering method
  is further used for finding the roots. For each fiber, the growth starts 
 from the root point and moves step by step until the ending, where each st
 ep is defined as an oriented line with a constant length according to the 
 predicted orientation field. To determine when to end, a pixel-aligned RNN
  architecture is designed to form a binary classifier, which outputs stop 
 or not for each growing step. To support the training of all proposed netw
 orks, we build the first 3D synthetic eyebrow dataset that contains 400 hi
 gh-quality eyebrow models manually created by artists. Extensive experimen
 ts have demonstrated the effectiveness of the proposed EMS pipeline on a v
 ariety of different eyebrow styles and lengths, ranging from short and spa
 rse to long bushy eyebrows.\n\nRegistration Category: Full Access\n\nSessi
 on Chair: Weidan Xiong (Shenzhen University)
URL:https://asia.siggraph.org/2023/full-program?id=papers_231&sess=sess137
END:VEVENT
END:VCALENDAR