Pixel-Light Attention for Neural Many-Light Importance Sampling
ACM Transactions on Graphics · 09.11.2026 · 45 (6) · 1-14
DOI: 10.1145/3842511 ↗Kurz gefasst
Fachartikel in der Zeitschrift ACM Transactions on Graphics, online erschienen am 09.11.2026. Verfasst von Xuejun Hu und 2 weiteren Autor:innen. Die Arbeit nennt 46 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China.
Abstract (Originalsprache)
We present a neural approach for many-light importance sampling built on a pixel-light attention mechanism for guided hierarchical traversal. Prior learning-based methods confine themselves to a single coarse level of the light hierarchy and entangle pixel and light information in a single learned object, limiting how finely the sampling distribution can adapt. Our attention scores each branch via a dot product between a learnable pixel feature and a direction-aware light node feature based on spherical-harmonic bases — factorizing the two sides into separate learnable representations while remaining lightweight enough to run at every traversal step. We train all features online during rendering by minimizing a KL-divergence loss against the ideal distribution, with gradients accumulated along each path via an explicit re-traversal pass. Our method achieves consistently lower error than state-of-the-art techniques across a range of challenging scenes.
Förderung
- National Natural Science Foundation of China
Zitieren
APA
Hu, X., Lu, J., & Xu, K. (2026). Pixel-Light Attention for Neural Many-Light Importance Sampling. ACM Transactions on Graphics, 45(6), 1-14. https://doi.org/10.1145/3842511
Vancouver
Hu X, Lu J, Xu K. Pixel-Light Attention for Neural Many-Light Importance Sampling. ACM Transactions on Graphics. 2026;45(6):1-14. doi:10.1145/3842511
BibTeX
@article{Hu2026,
author = {Hu, Xuejun and Lu, Jinfan and Xu, Kun},
title = {Pixel-Light Attention for Neural Many-Light Importance Sampling},
journal = {ACM Transactions on Graphics},
year = {2026},
volume = {45},
number = {6},
pages = {1--14},
doi = {10.1145/3842511}
}
EUVIMED