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:20240214T070241Z
LOCATION:Darling Harbour Theatre\, Level 2 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231212T093000
DTEND;TZID=Australia/Melbourne:20231212T124500
UID:siggraphasia_SIGGRAPH Asia 2023_sess209_papers_130@linklings.com
SUMMARY:GroomGen: A High-Quality Generative Hair Model Using Hierarchical 
 Latent Representations
DESCRIPTION:Technical Papers\n\nYuxiao Zhou (ETH Zürich), Menglei Chai and
  Alessandro Pepe (Google Inc.), Markus Gross (ETH Zürich), and Thabo Beele
 r (Google Inc.)\n\nDespite recent successes in hair acquisition that fits 
 a high-dimensional hair model to a specific input subject, generative hair
  models, which establish general embedding spaces for encoding, editing, a
 nd sampling diverse hairstyles, are way less explored. In this paper, we p
 resent GroomGen, the first generative model designed for hair geometry com
 posed of highly-detailed dense strands. Our approach is motivated by two k
 ey ideas. First, we construct hair latent spaces covering both individual 
 strands and hairstyles. The latent spaces are compact, expressive, and wel
 l-constrained for high-quality and diverse sampling. Second, we adopt a hi
 erarchical hair representation that parameterizes a complete hair model to
  three levels: single strands, sparse guide hairs, and complete dense hair
 s. This representation is critical to the compactness of latent spaces, th
 e robustness of training, and the efficiency of inference. Based on this h
 ierarchical latent representation, our proposed pipeline consists of a str
 and-VAE and a hairstyle-VAE that encode an individual strand and a set of 
 guide hairs to their respective latent spaces, a hybrid densification step
  that populates sparse guide hairs to a dense hair model, and a neural sim
 ulator that deforms hair driven by head pose. GroomGen not only enables no
 vel hairstyle sampling and plausible hairstyle interpolation, but could al
 so enable interactive editing of complex hairstyles, or serve as strong da
 ta-driven prior for hairstyle reconstruction from images. We demonstrate t
 he superiority of our approach with qualitative examples of diverse sample
 d hairstyles and quantitative evaluation of generation quality regarding e
 very single component and the entire pipeline.\n\nRegistration Category: F
 ull Access, Enhanced Access, Trade Exhibitor, Experience Hall Exhibitor
URL:https://asia.siggraph.org/2023/full-program?id=papers_130&sess=sess209
END:VEVENT
END:VCALENDAR