Presentation

<
Towards Comprehensive Neural Materials: Dynamic Structure-Preserving Synthesis with Accurate Silhouette at Instant Inference Speed
SessionNeural Materials & LOD
DescriptionWe challenge the comprehensive neural material representation by thoroughly considering the essential aspects of the complete appearance. We introduce an int8-quantized model that keeps high fidelity while achieving an order of magnitude speedup compared to previous methods, and a controllable structure-preserving synthesis strategy, along with accurate displacement effects.
Authors
Zilin Xu
University of California Santa Barbara
Xiang Chen
Shandong University
Chen Liu
Zhejiang Lingdi Digital Technology Co.,Ltd
Beibei Wang
Nanjing University
Lu Wang
Shandong University
Zahra Montazeri
University of Manchester
Ling-Qi Yan
University of California Santa Barbara
Event Type
Technical Paper
TimeThursday, 14 August 20252:00pm - 2:10pm PDT
LocationWest Building, Rooms 301-305
ACM Digital Library Journal Papers' PDFs
Conference Papers' PDFs
Session Time & Location
Sunday, 10 August 20256:00pm - 8:45pm PDTWest Building, Ballroom AB
Thursday, 14 August 20252:00pm - 3:30pm PDTWest Building, Rooms 301-305
Recordings
Livestreamed
Recorded
Next PresentationNext PresentationGenerative Neural Materials
Similar Presentations
Appearance-Preserving Scene Aggregation for Level-of-Detail Rendering
Generative Neural Materials
Neural BRDF Importance Sampling by Reparameterization