Presentation

<
VideoNeuMat: Neural Material Extraction from Generative Video Models
SessionMaterials & Textures
DescriptionVideoNeuMat extracts reusable neural materials from video diffusion models. It fine-tunes a large video model as a virtual gonioreflectometer and uses a Large Reconstruction Model (LRM) to reconstruct compact neural materials in a single pass. The resulting materials are realistic, diverse, and generalize to novel viewing and lighting conditions.
Authors
Bowen Xue
University of Manchester
Saeed Hadadan
NVIDIA
Zheng Zeng
University of California Santa Barbara
NVIDIA
Fabrice Rousselle
NVIDIA
Zahra Montazeri
University of Manchester
Milos Hasan
NVIDIA
Event Type
Technical Paper
TimeTuesday, 21 July 202610:55am - 11:05am PDT
LocationRoom 408 A
Digital Library PDF
Session Time & Location
Sunday, 19 July 20266:00pm - 8:45pm PDTHall K
Tuesday, 21 July 202610:45am - 12:15pm PDTRoom 408 A
Keywords
Modeling
Rendering
Registration Categories
Full Conference Supporter
Full Conference
Next PresentationNext PresentationPureSample: Neural Materials Learned by Sampling Microgeometry
Similar Presentations
Toward Richer Material Generation via Procedural Data Enhancement
PureSample: Neural Materials Learned by Sampling Microgeometry
8DNA: 8D Neural Asset Light Transport by Distribution Learning