BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Asia/Tokyo
X-LIC-LOCATION:Asia/Tokyo
BEGIN:STANDARD
TZOFFSETFROM:+0900
TZOFFSETTO:+0900
TZNAME:JST
DTSTART:18871231T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20250110T023312Z
LOCATION:Hall B5 (2)\, B Block\, Level 5
DTSTART;TZID=Asia/Tokyo:20241204T110800
DTEND;TZID=Asia/Tokyo:20241204T111900
UID:siggraphasia_SIGGRAPH Asia 2024_sess113_papers_253@linklings.com
SUMMARY:Cafca: High-quality Novel View Synthesis of Expressive Faces from 
 Casual Few-shot Captures
DESCRIPTION:Technical Papers\n\nMarcel C. Buehler and Gengyan Li (ETH Züri
 ch, Google VR); Erroll Wood, Leonhard Helminger, Xu Chen, Tanmay Shah, Dao
 ye Wang, Stephan Garbin, and Sergio Orts Escolano (Google VR); Otmar Hilli
 ges (ETH Zürich); and Dmitry Lagun, Jérémy Riviere, Paulo Gotardo, Thabo B
 eeler, Abhimitra Meka, and Kripasindhu Sarkar (Google VR)\n\nVolumetric mo
 deling and neural radiance field representations have revolutionized 3D fa
 ce capture and photorealistic novel view synthesis. However, these methods
  often require hundreds of multi-view input images and are thus inapplicab
 le to cases with less than a handful of inputs.\nWe present a novel volume
 tric prior on human faces that allows for high-fidelity expressive face mo
 deling from as few as three input views captured in the wild. Our key insi
 ght is that an implicit prior trained on synthetic data alone can generali
 ze to extremely challenging real-world identities and expressions and rend
 er novel views with fine idiosyncratic details like wrinkles and eyelashes
 .\nWe leverage a 3D Morphable Face Model to synthesize a large training se
 t, rendering each identity with different expressions, hair, clothing, and
  other assets. We then train a conditional Neural Radiance Field prior on 
 this synthetic dataset and, at inference time, fine-tune the model on a ve
 ry sparse set of real images of a single subject. On average, the fine-tun
 ing requires only three inputs to cross the synthetic-to-real domain gap. 
 The resulting personalized 3D model reconstructs strong idiosyncratic faci
 al expressions and outperforms the state-of-the-art in high-quality novel 
 view synthesis of faces from sparse inputs in terms of perceptual and phot
 o-metric quality.\n\nRegistration Category: Full Access, Full Access Suppo
 rter\n\nLanguage Format: English Language\n\nSession Chair: Forrester Cole
  (Google)
URL:https://asia.siggraph.org/2024/program/?id=papers_253&sess=sess113
END:VEVENT
END:VCALENDAR