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:20250110T023309Z
LOCATION:Hall B5 (2)\, B Block\, Level 5
DTSTART;TZID=Asia/Tokyo:20241203T151900
DTEND;TZID=Asia/Tokyo:20241203T153100
UID:siggraphasia_SIGGRAPH Asia 2024_sess107_papers_162@linklings.com
SUMMARY:Reflection-Aware Neural Radiance Fields
DESCRIPTION:Technical Papers\n\nChen Gao, Yipeng Wang, and Changil Kim (Me
 ta); Jia-Bin Huang (University of Maryland, College Park); and Johannes Ko
 pf (Meta)\n\nNeural Radiance Fields (NeRF) have demonstrated exceptional c
 apabilities in reconstructing complex scenes with high fidelity. However, 
 NeRF's view dependency can only handle low-frequency reflections. It falls
  short when handling complex planar reflections, often interpreting them a
 s erroneous scene geometries and leading to duplicated and inaccurate scen
 e representations. To address this challenge, we introduce a reflection-aw
 are NeRF that jointly models planar reflectors, such as windows, and expli
 citly casts reflected rays to capture the source of the high-frequency ref
 lections. We query a single radiance field to render the primary color and
  the source of the reflection. We propose a sparse edge regularization to 
 help utilize the true sources of reflections for rendering planar reflecti
 ons rather than creating a duplicate along the primary ray at the same dep
 th. As a result, we obtain accurate scene geometry. Rendering along the pr
 imary ray results in a clean, reflection-free view, while explicitly rende
 ring along the reflected ray allows us to reconstruct highly detailed refl
 ections. Our extensive quantitative and qualitative evaluations of real-wo
 rld datasets demonstrate our method's enhanced performance in accurately h
 andling reflections.\n\nRegistration Category: Full Access, Full Access Su
 pporter\n\nLanguage Format: English Language\n\nSession Chair: Hongzhi Wu 
 (Zhejiang University; State Key Laboratory of CAD&CG, Zhejiang University)
URL:https://asia.siggraph.org/2024/program/?id=papers_162&sess=sess107
END:VEVENT
END:VCALENDAR