A Real‐Time Multi‐Scale Neural Representation for Complex Surface Reflectance
Computer Graphics Forum · 2026-04-21
DOI: 10.1111/cgf.70331 ↗At a glance
Journal article in Computer Graphics Forum, published online on 2026-04-21. Written by Heikki Timonen and 2 co-authors. Cited 1 times so far (Crossref); 86 references. The full text is freely available (open access, CC BY 4.0). Funded by: European Research Council.
Abstract (original language)
Recent machine learning methods have significantly advanced the state of the art in the classic problem of representing surface appearance over angle, space, and scale. The models tend, however, to be relatively heavy compared to traditional fixed‐function representations, making real‐time application challenging. We present a neural shading architecture that allows the use of smaller and faster‐to‐evaluate neural networks than current state of the art, while faithfully representing complex spatial and angular variation. We target the angular complexity that arises both from prefiltering normal‐mapped SVBRDFs, as well as complex, measured homogeneous BRDFs. A key architectural innovation is the introduction of a multiplicative interaction (“gating”) between learnable parameters that significantly increases our model's expressive power. Our straightforward, unop‐timized shader implementation renders over 1000 full HD frames per second on a consumer GPU using our default parameters.
Funding
- European Research Council
Cite
APA
Timonen, H., Kemppinen, P., & Lehtinen, J. (2026). A Real‐Time Multi‐Scale Neural Representation for Complex Surface Reflectance. Computer Graphics Forum. https://doi.org/10.1111/cgf.70331
Vancouver
Timonen H, Kemppinen P, Lehtinen J. A Real‐Time Multi‐Scale Neural Representation for Complex Surface Reflectance. Computer Graphics Forum. 2026. doi:10.1111/cgf.70331
BibTeX
@article{Timonen2026,
author = {Timonen, Heikki and Kemppinen, Pauli and Lehtinen, Jaakko},
title = {A Real‐Time Multi‐Scale Neural Representation for Complex Surface Reflectance},
journal = {Computer Graphics Forum},
year = {2026},
doi = {10.1111/cgf.70331}
}
EUVIMED