Presentation / Installation


NeRO: Neural Geometry and BRDF Reconstruction of Reflective Objects From Multiview Images
SessionNeural Capturing
DescriptionIn this paper, we address the challenging problem of multiview reconstruction of reflective objects in an unknown environment. We present a neural rendering-based method called NeRO, which enables not only high-fidelity shape reconstruction but also accurate BRDF estimation with photorealistic relighting.
Authors
Yuan Liu
University of Hong Kong
Peng Wang
University of Hong Kong
Cheng Lin
Tencent Games
Xiaoxiao Long
University of Hong Kong
Jiepeng Wang
University of Hong Kong
Lingjie Liu
Max Planck Institute for Informatics
University of Pennsylvania
Taku Komura
University of Hong Kong
Wenping Wang
Texas A&M University
Event Type
Technical Paper
TimeWednesday, 9 August 20233:56pm - 4:07pm PDT
LocationPetree Hall C
ACM Digital Library Technical Papers pdfs
Session Time & Location
Sunday, 6 August 20236pm - 8:30pm PDTWest Hall B
Wednesday, 9 August 20233:45pm - 5:15pm PDTPetree Hall C
Interest Areas
Research & Education
Recordings
Livestreamed
Recorded
Keywords
Rendering
Video
Registration Categories
FC
FCS
V
VS
EFC
Next PresentationNext Presentation
BakedSDF: Meshing Neural SDFs for Real-time View Synthesis